The decisions a joint-venture agreement has to name before anyone funds

Most joint-venture negotiations spend their energy on the split. That matters, but the split is only the distribution of an outcome the partners have to reach together first. Governance is what determines whether they can. An agreement with a fair waterfall and no decision framework is a good deal that cannot be executed once conditions change — and conditions always change.

The six categories below are the ones we would not sign without. For each, the agreement needs three things: who holds the decision, what threshold applies, and what happens if the threshold is not met.

1. The budget, and amendments to it

An approved budget is the operating licence of the managing party. Without one, every expenditure is arguable; with one, only departures are. The useful mechanics are a defined annual or project budget, a stated variance tolerance by line and in aggregate, and a clear rule on whether savings in one line may fund overruns in another. State plainly what happens if a new budget is not approved: continuing on the prior budget adjusted for contractual commitments is a workable default, and far better than silence.

2. Capital calls, and the consequence of not funding

This is the single most consequential clause in a joint-venture agreement, because it is the one that changes ownership. It needs the trigger, the notice period, the funding window, and the remedy — dilution on a stated formula, a member loan at a stated rate and priority, or a preferred return step-up. Whatever the remedy, both parties should be able to calculate the outcome of a missed call before they sign, not discover it under stress.

• Is a capital call limited to the approved budget and cost overruns, or discretionary?
• Is there a cap on total calls, or a point at which a non-funding party's exposure is fixed?
• Does the dilution formula use committed capital, contributed capital or appraised value — the three give materially different answers.

3. Financing and encumbrance

Placing debt on the asset, refinancing it, granting security, and giving guarantees. Guarantees deserve their own treatment: who signs, how guarantee exposure is compensated or shared, and whether a guarantor's consent is required for anything that increases their exposure. A partner carrying recourse on a loan and holding no consent right over its amendment is in a position no amount of goodwill fixes.

4. Sale and refinance timing

Partners commonly agree on the strategy and disagree on the date. Mechanics that work: a stated hold period, a marketing-decision threshold, a right of first offer between partners, and a forced-sale route after a defined date. A drag-along needs a matching tag-along, or the minority is exposed to a transaction it cannot join.

5. Replacement of the managing party

Removal for cause should be defined by conduct, not by disappointment: fraud, wilful misconduct, material uncured breach, insolvency, criminal conviction, loss of a required licence, unauthorised transfer. Removal without cause, if the agreement permits it, needs a threshold and an economic consequence, because the managing party's promote is usually the reason it accepted the role. State what happens to that interest on each type of removal, and who steps in — a removal clause with no named successor manager stops a project rather than fixing it.

6. Material changes to the business plan

Change of use, change of unit mix or product, phasing changes, a materially different delivery method, or a change in exit strategy. These are the decisions that convert the deal a partner underwrote into a different deal. Defining 'material' with thresholds — a percentage of budget, a percentage of units, a change in entitled use — is better than reserving 'material changes' generally and arguing about the adjective later.

Deadlock: have a route, and prefer the boring one

Every reserved-matter list creates the possibility of deadlock. The routes in common use, roughly in order of how destructive they are: escalation to named principals within a fixed period, an independent expert or appraiser for valuation questions, mediation, a buy-sell mechanism, and forced sale to a third party. Escalation and expert determination resolve most real disagreements. A buy-sell mechanism is a genuine remedy but favours the partner with liquidity, which is worth understanding before it is the only clause left.

One practical test before signing: take three plausible bad scenarios — a cost overrun requiring capital neither partner planned, a lender requiring a guarantee extension, and an offer to sell two years early — and read the agreement to find who decides each. If you cannot point to the clause in a few minutes, the agreement is not finished.

Joint-venture agreements are legal instruments with tax consequences. This article describes practice, not advice, and it is not a substitute for your counsel and tax adviser drafting and reviewing the documents for your transaction.