The direct answer
With limited capital the question is not how much you need. It is what you buy first, and what you deliberately do not buy yet.
The order that follows is the one the evidence supports.
First, secure what cannot be undone. Tenure and access. Money spent improving land you do not securely hold is money at risk, and no later step recovers it.
Second, find out what you have. Soil, water, slope, access. Diagnosis before purchase. This is the step most often skipped and the one with the clearest published justification, set out below.
Third, buy the things that determine whether the crop or the enterprise exists at all, in the order the season imposes: seed or stock of a known class, and the labour to establish it on time.
Fourth, buy the things that protect what you have already spent. Weeding within the timing window, basic storage, basic safety provision.
Fifth, and only fifth, buy the things that raise output on a base that is already working. Fertiliser where a soil test justifies it, irrigation, machinery, structures.
The single most useful piece of published evidence for that ordering is uncomfortable. A 2019 study of fertiliser response across African smallholder maize farms, reviewing 71 on-farm publications from Kenya and sub-Saharan Africa spanning 1986 to 2016, found that "In SSA, 18% of the plots were non-responsive plots with an FR < 1." A fertiliser response ratio below one means the fertilised plot did not out-yield the unfertilised one. On roughly one plot in five in that review, buying fertiliser returned less than nothing. The same study notes that "Currently, blanket recommendations are provided across agro-ecological zones, although fertilizer response and nutrient use efficiency by maize crop are spatially variable."
For someone with limited capital, that is the whole argument for diagnosis before purchase, in one number.
This article publishes no start-up cost figure, no enterprise budget and no income, profit or payback claim. The section below explains why, and it is not a hedge.
Who this article is for
You are starting a farm in Uganda, or expanding one, and money is the binding constraint. You may have inherited or rented land, or be about to. You have been told several different figures for what farming costs to start, none of them sourced. You want to know what to do with the money you actually have, in what order.
This is also for you if you already farm at subsistence scale and are deciding what the first commercial step should be.
The problem this article solves
The invented-number problem. Search for what farming costs to start in Uganda and you will find confident figures. Falcon looked for a credible, current, primary-source enterprise budget or start-up cost figure for Ugandan smallholder agriculture and could not verify one. The only current cost breakdowns located were social media posts. The best Ugandan input expenditure data found was published in 2010 using 2005/06 survey data, which is around twenty years old and useless as a budget. So the honest position is that the numbers circulating are not sourced, and a plan built on them is built on nothing.
The wrong-order problem. People with limited capital buy inputs first, because inputs are the visible, purchasable, hopeful thing. The fertiliser response evidence above says that is exactly the wrong order when you have not diagnosed the field.
The all-in problem. Limited capital creates pressure to commit everything to one season, because splitting it feels like half-measures. But the Ugandan drought record shows consecutive reduced seasons, and a farm that commits everything to one season has nothing left to farm the next one with.
The scale problem. Most Ugandan agricultural households are working with small areas. The national annual agricultural survey for 2019 reported that average holding size reached 1.3 hectares per household and that 67 per cent of agricultural households operate less than 1 hectare of agricultural land. Advice written for larger farms does not transfer.
The decision you are making
- What order will your available money be spent in, and what is the trigger that releases each next step?
- What are you deliberately deferring, and until when?
- What are you keeping back so that a bad season does not end the enterprise?
- Which constraints are gates you cannot pass, and which are things you can grow into?
You are not deciding what income you will earn. Nothing in this article predicts that, and anything that does should be treated with suspicion.
Uganda-specific context
The scale most Ugandan farming actually happens at
The 2019 annual agricultural survey reported that average holding size increased over the previous decade to an average of 1.3 hectares per household, up from 1.1 hectares in 2008/09, and that 67 per cent of agricultural households operate less than 1 hectare of agricultural land.
The national household survey for 2019/20 reported that "39 percent of households (3.5 million) were in the subsistence economy compared to 61 percent (5.4 million) in the non-subsistence economy in 2019/20", that of subsistence households "62 percent were engaged mainly in subsistence agriculture", and that "The percentage of the population in poverty reduced by 1.1 percentage points between the survey periods, that is, from 21.4 percent in 2016/17 to 20.3 percent in 2019/20."
Those figures are from the 2019 and 2019/20 survey rounds and are quoted with those dates. They set the context: the typical Ugandan farm start is on a small area, alongside subsistence production, with limited cash.
The Parish Development Model, and what it actually is
The Parish Development Model is described by the Ministry of Local Government as "A multi-sectoral strategy to create socio-economic transformation strategy by moving the 39% households out of the subsistence economy into the money economy", with the objective to "Increase Household incomes and improve quality of life of Ugandans with a specific focus on the total economic transformation of the Households through getting Ugandans out of the subsistence economy into the money economy". Coordination sits with the PDM Secretariat: "The PDM Secretariat is the coordinating entity for the Parish Development Model countrywide. Domiciled at the Ministry of Local Government, it is headed by a PDM Coordinator and deputy coordinator."
Its seven pillars, as published, are: "Agricultural Value-Chain Development (Production, Storage, Processing and Marketing)"; "Infrastructure and Economic Services"; "Financial Inclusion"; "Social Services"; "Community Mobilisation and Mind-set change"; "Parish-Based Management Information System"; and "Governance and Administration".
The financial inclusion pillar is the one that matters to a farm start. The user handbook for the Parish Revolving Fund under pillar three, published in October 2022, states that "Government shall provide a conditional Grant, of UGX100million per annum, to each of the gazetted 10,594 parishes in Uganda", that "This amount, including charges and insurance, shall not exceed 1million shillings per household", and that "For the Financial Year 2022/23, this rate shall be fixed at six percent (6%) per annum which translates to 0.5% per month." It states that "The fund shall be used for the sole purpose of lending, by the PDM SACCO, to subsistence households in the parish that have identified (and want to run) viable agri-based income generating activities/enterprises", and gives eligibility as households that are "members of a PDM Enterprise Group that is a member of the PDM SACCO" and that "have a running enterprise and/or a viable business plan."
Three cautions, all important.
The rate quoted is expressly stated for the financial year 2022/23. Do not treat it as the current rate. Ask your parish or sub-county whether it still applies and what the current terms are.
The programme lends. Money that arrives through a PDM SACCO is a loan to be repaid, and it should be treated with exactly the discipline you would apply to any other borrowing.
Eligibility runs through group and SACCO membership and requires a running enterprise or a viable business plan. That means the sequencing in this article and the plan document owned by the business plan article referenced at the end are the work you do before, not after, you approach the fund.
What Falcon could not verify, and why it matters here more than anywhere
Falcon could not verify any current, credible, primary-source enterprise budget or start-up cost figure for Ugandan smallholder agriculture. Not for maize, not for poultry, not for horticulture, not for any enterprise. The only current cost breakdowns located during this research were social media posts with no stated method, which cannot be cited.
The best Ugandan input expenditure data located came from a 2010 research paper using 2005/06 national household survey data. It reports the relative weight of hired labour against all other inputs combined, which is a structurally useful finding, but its shilling amounts are around twenty years old and inflation and exchange rate movement make them worthless as a budget. No shilling figure from that source is reproduced in this article.
The structural finding is worth keeping even though the amounts are not: in that data, hired labour cost more than all other inputs combined. If the structure still holds, then labour, not inputs, is the largest cash requirement in a Ugandan crop enterprise, and the labour article referenced at the end is where the planning for it belongs. Treat that as a hypothesis to test against your own costs, not as a current fact.
What to do instead of using a published figure. Build your own budget from local prices. Walk to your trading centre, ask what seed, labour, transport and any input actually cost this season, write those numbers down with the date, and build from there. That takes a morning and it produces a budget that is true for your district in this season, which no published national figure can be.
Step by step: the sequence
Step 1: settle what you are farming, before you spend anything
Enterprise selection comes first and it is owned by the enterprise selection article referenced at the end. If you have not done that work, do it before reading further, because everything below assumes you know what you are starting.
Step 2: secure tenure and access, and treat it as a gate
Do not improve land you do not securely hold. The land title and tenure article referenced at the end covers how to verify the position. This is a gate, not a step you can defer: money spent on a boundary you cannot defend, a structure on land you do not hold, or a perennial crop on land you may lose, is money that cannot be recovered.
Access matters alongside tenure. Land you cannot reach in the wet season, or from which you cannot move produce, imposes a cost every season.
Step 3: diagnose before you buy
Soil test. Understand your water situation. Look honestly at slope and drainage. The soil testing article, the water sources article, the erosion control article and the flooding article referenced at the end all cover the diagnosis side.
The justification is the fertiliser response evidence quoted at the top of this article. When roughly one plot in five in a large review did not respond to fertiliser at all, and when that review found that available data on soil, climate and management explained less than a third of the variation it observed, buying inputs before you know your field is a bet at odds you cannot see. The cost of a soil test is small relative to the cost of a fertiliser purchase that returns nothing.
Diagnosis also tells you which constraint is actually binding. If water is the binding constraint, fertiliser will not fix it. If drainage is the binding constraint, better seed will not fix it. Spending on the non-binding constraint is the most common way limited capital disappears.
Step 4: buy what determines whether the enterprise exists
For a crop, this is seed of a known class, planted on time, at a spacing you can execute. For livestock, it is stock of known origin and health status. For either, it is the labour required to establish on time, because a late establishment costs more than most inputs save.
Buy seed by class and by source, not by price alone. The maize article referenced at the end deals with variety verification and seed classes.
Step 5: protect what you have already spent
This is the cheapest high-return spending on any farm and it is routinely deferred.
- Weeding within the timing window. The weeding article referenced at the end explains why the timing of the first weeding decides more than the total number of weedings.
- Basic storage, so that what you harvest is not lost between the field and the sale.
- Basic safety provision, covered in the farm safety article. A serious injury is a capital event on a small farm.
- A record book, which costs almost nothing and is what makes every subsequent decision evidence-based.
Step 6: keep something back
Do not commit everything to one season. The Ugandan record documented in the drought article referenced at the end shows consecutive reduced seasons, not isolated bad years. What you keep back is seed for next season, breeding stock, and enough cash to plant again.
Decide the reserve as a proportion before the season starts, and treat it as unavailable. A reserve that is available is a reserve that gets spent.
Step 7: defer deliberately, with a written trigger
Deferral is a decision, not a failure. Write down what you are deferring and what will release it: a completed season, a specific record, a specific price achieved, a repaid loan. A deferred item without a trigger becomes either a permanent gap or an impulse purchase.
Step 8: measure, then step up
Each season, record what was spent, what was produced and what was received. That record is what converts the next spending decision from hope into evidence, and it is also what the business plan article's plan document and any lender will want.
Decision framework: what to buy now and what to defer
| Category | Position in the sequence | Why | Defer until |
|---|---|---|---|
| Tenure and access | Gate, before anything | Irrecoverable if wrong | Never defer. This is a condition, not a purchase |
| Diagnosis: soil test, water, slope, drainage | First spending | Determines whether every later purchase works | Never defer. It is small relative to what it protects |
| Seed or stock of known class | Core | Determines whether there is an enterprise at all | Never defer below what one workable area requires |
| Establishment labour, on time | Core | Late establishment cannot be recovered by later spending | Never defer |
| Weeding within the timing window | Protective | Protects the spending already made | Never defer |
| Basic storage | Protective | Prevents loss of what has already been produced | Defer scale, not existence |
| Basic safety provision | Protective | An injury is a capital event on a small farm | Never defer |
| Records | Protective | Makes every later decision evidence-based | Never defer. Cost is negligible |
| Fertiliser | Growth, conditional | Around one plot in five in a large review did not respond | Until a soil test justifies it |
| Irrigation | Growth | High cost, and only pays where water is the binding constraint | Until the constraint is confirmed and the base enterprise works |
| Machinery | Growth | Usually cheaper to hire until scale justifies owning | Until the volume of work justifies it against hire cost |
| Permanent structures | Growth | Long payback, and requires secure tenure | Until tenure and the enterprise are both settled |
| Expansion of area | Growth | Expanding a system that does not yet work multiplies a loss | Until one area works and is recorded as working |
| Perennial crops | Growth | Long commitment on land and on capital | Until tenure is secure and cash flow can bridge the establishment period |
The right-hand column is the article. Anyone can list what a farm needs. The value is in stating what is deferred and what releases it.
Field record: the sequencing file
| Section | What it holds | Updated |
|---|---|---|
| Enterprise decision | What you are farming and why, cross-referenced to the enterprise selection work | At the start, and at any change |
| Tenure position | What you hold, how, what was verified, when and by whom | At the start, and at any change |
| Diagnosis record | Soil test date and results, water situation, slope and drainage findings | At the start, and at each retest |
| Spending sequence | Item, category, planned position in the sequence, actual date bought, amount | Continuously |
| Deferral list | Item deferred, reason, and the written trigger that releases it | At each review |
| Reserve | The proportion held back, in what form, and whether it was drawn on | Each season |
| Local price notes | What seed, labour, transport and inputs actually cost, with the date and the source | Each season |
| Borrowing record | Source, amount, terms, rate, repayment schedule, and the date the terms were confirmed | On each borrowing |
| Season outcome | Spent, produced, received, and what was learned | Each season |
The local price notes row is what replaces the published start-up cost figure that does not exist. After two seasons it is better than any national average would have been.
Cost categories
Falcon publishes no amounts. No credible current Ugandan start-up cost or enterprise budget figure was verified during this research, and the article will not invent one. Price each of these locally, with the date, and build your own budget.
- Tenure verification and any associated fees.
- Soil testing.
- Land preparation.
- Seed or stock.
- Establishment labour.
- Weeding labour, which the structure of older Ugandan data suggests may be the largest single cash item.
- Basic storage.
- Basic safety provision, including protective equipment where chemicals are used.
- Transport to market.
- A record book and a rain gauge.
- Deferred items, priced but not bought, so you know what the next step costs.
- The reserve, held rather than spent.
Common mistakes
- Starting from a start-up cost figure found online. No verified current Ugandan figure exists publicly, so the figure came from somewhere unsourced.
- Buying inputs before diagnosis, when review evidence found around 18 per cent of sub-Saharan plots did not respond to fertiliser at all.
- Spending on the constraint that is not binding: fertiliser where water is the problem, or seed where drainage is the problem.
- Improving land whose tenure is not secure.
- Committing everything to one season, when the Ugandan record shows consecutive reduced seasons.
- Expanding area before one area works and has been recorded as working.
- Buying machinery before the volume of work justifies owning rather than hiring.
- Planting perennials before tenure is secure and before cash flow can bridge establishment.
- Treating PDM money as a grant to the household rather than as a loan to be repaid.
- Quoting the 6 per cent rate without checking, when the handbook states it for the financial year 2022/23.
- Deferring items without writing down the trigger that releases them, so deferral becomes drift.
- Keeping no records, so that every season's spending decision repeats the last one's guess.
- Underestimating labour, which older Ugandan data suggests exceeded all other inputs combined.
- Deferring safety provision, when an injury on a small farm is a capital event.
Warning signs and escalation points
| What you observe | What it means | What to do |
|---|---|---|
| You cannot state your tenure position precisely | You are about to spend on land you may not securely hold | Stop and verify before any further spending |
| You are about to buy fertiliser without a soil test | You are betting at odds you cannot see | Test first; the cost is small relative to the purchase |
| A cost figure you are relying on has no stated source | It is very likely invented | Price it locally yourself, with the date |
| Your whole available capital is committed to one season | A single reduced season ends the enterprise | Set a reserve as a proportion and treat it as unavailable |
| You are expanding area while the first area is not yet working | You are multiplying an unproven system | Make one area work and record it before expanding |
| You are considering machinery | Ownership rarely beats hire below a volume threshold | Compare against local hire cost for your actual volume |
| A lender or programme is offering money quickly | Speed is not a term. Rate, schedule and consequences are | Get the terms in writing and confirm the current rate |
| The PDM terms you were told differ from the handbook | The handbook rate was stated for FY2022/23 and terms may have changed | Confirm current terms with your parish or sub-county |
| Labour costs are larger than you budgeted | This is consistent with the structure of older Ugandan data | Rebuild the budget around labour, using the labour planning article |
| You are deferring safety provision to fund an input | An injury is a capital event on a small farm | Do not. Safety provision sits in the protective tier |
When this approach works and when it does not
It works when enterprise selection has been done, when tenure is settled, when diagnosis precedes purchase, and when deferrals have written triggers.
It works particularly well at small scale, because sequencing matters most when there is no slack.
It does not work if the enterprise itself is wrong for the land, the climate, the market or the operator. Sequencing spending on a badly chosen enterprise makes the loss slower, not smaller.
It does not work if the reserve is treated as available. A reserve exists to fund the next season, and its whole value is that it is not spent this one.
It does not predict income. Nothing in this article states what you will earn, and no responsible source can, because output depends on rainfall, prices and conditions nobody controls.
It is not financial advice. Borrowing decisions, including under a government programme, have consequences that depend on your own circumstances.
A low-risk first step
Spend one morning at your nearest trading centre with a notebook, and write down, with today's date, what these actually cost right now: a kilogram of the seed you intend to plant, a day of weeding labour, transport of a bag to market, and a soil test through your district production office.
Four numbers, sourced, dated and local. That is a better foundation for a plan than any published figure, and it costs you a morning.
Then book the soil test before you buy anything else.
Limitations and professional-support triggers
Get professional support in these situations.
- Tenure that is unclear, disputed, customary or inherited without documentation. Take advice before spending; the land title and tenure article referenced at the end is the starting point.
- Any borrowing decision, including under the Parish Development Model. Confirm the terms in writing and understand the repayment obligation before signing.
- Interpreting a soil test. The soil testing article, and a qualified agronomist.
- Building a plan for a lender. The business plan article referenced at the end.
- Formalising the business. The registration and formalisation article referenced at the end.
Four verification gaps in this article are disclosed rather than hidden.
First, no current, credible, primary-source Ugandan start-up cost figure or enterprise budget was verified. The only current breakdowns located were social media posts, which cannot be cited. No amount is published here.
Second, the best Ugandan input expenditure data located is from a 2010 publication using 2005/06 data. Its structural finding that hired labour exceeded all other inputs combined is quoted as structure only, and none of its shilling amounts is reproduced, because they are around twenty years old.
Third, the Parish Revolving Fund interest rate quoted is stated in its source for the financial year 2022/23 and must be re-checked before it is relied on.
Fourth, no credible dedicated study on why new agricultural ventures fail in Africa was verified, so no failure statistic appears in this article and the sequencing argument rests on the fertiliser response evidence instead.
The survey figures quoted are from the 2019 annual agricultural survey and the 2019/20 national household survey and carry those dates. National statistics are revised.
Clear next actions
- Complete the enterprise selection work first, using the enterprise selection article, if you have not already.
- Verify your tenure position and treat it as a gate before any spending on the land.
- Book a soil test and establish your water, slope and drainage position before buying any input.
- Spend a morning collecting four local, dated prices: seed, a day of weeding labour, transport to market, and the soil test.
- Write the spending sequence for your own farm using the decision framework table, and write the trigger beside every deferred item.
- Set the reserve as a proportion of your capital, decide what form it is held in, and treat it as unavailable.
- If you intend to approach the Parish Development Model, confirm current terms with your parish or sub-county, and understand that it is a loan.
- Start the record book this season, including local prices with dates, so that next season's decisions are based on your own evidence.
Author
Falcon Foundation Ltd editorial team. Falcon Foundation Ltd is a privately owned Ugandan agricultural company established in 2013, registered in Uganda and operating land in Uganda across greenhouse crops, poultry, sheep, goats and beekeeping. This article sets out the order in which limited capital should be spent on a farm start and what should be deliberately deferred. It is general information only. It makes no income, profit, wealth or success promise, and it publishes no start-up cost figure or enterprise budget, because no current credible Ugandan source for one could be verified. It is not financial or legal advice. Confirm any programme terms with the responsible authority and take advice before borrowing.
