---
title: "Risk allocation and bankability | Optimize All Academy"
description: "What \"bankable\" means A project is bankable when lenders are willing to finance it on acceptable terms. Bankability is not a single test; it is the…"
url: https://optimizeall.com/learn/project-finance-and-financial-modelling/risk-allocation-and-bankability
updated: 2026-10-05
---

Project Finance & Financial Modelling · Capital structure, risk allocation and bankability · lesson 5 of 20 · 15 min

# Risk allocation and bankability

## What "bankable" means

A project is **bankable** when lenders are willing to finance it on acceptable terms. Bankability is not a single test; it is the combined result of predictable cash flows, sensible risk allocation, creditworthy counterparties, robust contracts, enforceable security and a sound legal framework.

## The core principle of risk allocation

**Allocate each risk to the party best able to manage it, at the lowest cost.** Risks that nobody can manage (e.g., some natural catastrophes) are insured, shared, or retained with appropriate pricing.

## The risk allocation matrix

| Risk | Typically borne by | Mitigation mechanism |
|---|---|---|
| Construction cost overrun | EPC contractor | Fixed-price, date-certain EPC contract |
| Construction delay | EPC contractor | Delay liquidated damages (LDs) sized to cover debt service and fixed costs |
| Performance shortfall | EPC contractor / O&M operator | Performance LDs, guarantees, warranties |
| Demand/volume | Offtaker or grantor (in contracted models) or SPV (merchant) | Take-or-pay PPA, availability payments, minimum revenue guarantees |
| Price | Offtaker or market | Fixed or indexed tariffs |
| Operating cost | O&M operator (partly) / SPV | Fixed-price O&M contract, indexation |
| Input/fuel supply | Supplier / offtaker | Long-term supply agreements, pass-through clauses |
| Interest rate | SPV | Hedging |
| Currency | SPV / offtaker | Tariff indexation, local currency debt, hedging |
| Change in law | Grantor/offtaker (often) | Compensation clauses |
| Political risk | Shared | Political risk insurance, DFI participation, government support |
| Force majeure | Shared | Insurance, relief and termination provisions |

## Key contract features lenders examine

- **EPC contract:** fixed price, fixed completion date, single-point responsibility, LDs, performance guarantees, security (performance bonds, parent guarantees), caps on liability that are not too low.
- **Offtake contract:** creditworthy offtaker, tenor at least as long as the debt, clear tariff formula, termination payments sufficient to repay debt in most scenarios.
- **O&M contract:** experienced operator, performance incentives and penalties, alignment with EPC warranties.
- **Direct agreements:** lenders' rights to step in and cure defaults before key contracts are terminated.
- **Security package:** mortgages/charges over assets, assignment of contracts and insurances, share pledges, control of project accounts.

## Back-to-back and gaps

The SPV must not be left holding risks it cannot bear. Check **back-to-back** alignment: if the offtake contract penalises the SPV for late commercial operation, the EPC delay LDs should cover at least that penalty plus debt service and fixed costs during the delay. Gaps (interface risk) are among the most common bankability problems.

## Worked example: a bankability review

*Illustrative.* A fictional 100 MW wind project in Sindh is presented to lenders. The review finds:

| Area | Finding | Lender concern | Fix |
|---|---|---|---|
| Offtake | 20-year PPA with state buyer; payment history includes delays | Payment risk | Payment security mechanism, sovereign support or DFI participation |
| Currency | Tariff partly indexed to USD; debt partly USD | Residual mismatch | Increase indexation or add local-currency tranche |
| EPC | Delay LDs capped at 10% of contract price | May be insufficient for long delays | Review cap vs expected delay costs; add sponsor completion support |
| Wind resource | Single year of on-site data | Energy yield uncertainty | Longer data, independent yield assessment, size debt on conservative (e.g., P90) yield |
| Grid | Connection timetable depends on utility | Interface risk | Deemed energy payments if grid not ready |

After these changes, lenders agree to finance, but at a lower gearing than first proposed, reflecting residual risk.

## Energy yield and P50/P90

For renewables, lenders usually size debt on a conservative production estimate (often a P90 or P99 one-year or ten-year case) while equity may use the P50 case. "P90" means production expected to be exceeded with 90% probability. The same logic applies to traffic forecasts in toll roads: lenders use downside cases.

## Environmental and social standards

Many lenders, especially DFIs and banks that have adopted the Equator Principles, require compliance with international environmental and social standards (for example IFC Performance Standards). Non-compliance can make a project unbankable regardless of its economics.

## Common mistakes

- Assuming a contract transfers risk when its liability caps make the transfer meaningless.
- Offtake tenor shorter than debt tenor.
- Ignoring interface risk between EPC, O&M and offtake.
- Underestimating the importance of counterparty creditworthiness.
- Leaving environmental and social assessments too late.

## Hands-on: residual-exposure checks in Excel

```text
Delay gap per day        =Offtake_penalty_per_day + Debt_service_per_day + Fixed_costs_per_day - EPC_LD_per_day
Days until LD cap        =EPC_LD_cap / EPC_LD_per_day
Uncovered exposure for N days
                         =IF(N<=Days_until_cap, N*Delay_gap,
                             Days_until_cap*Delay_gap + (N-Days_until_cap)*(Offtake_penalty_per_day+Debt_service_per_day+Fixed_costs_per_day))
Currency mismatch share  =USD_debt_service_share - USD_indexed_tariff_share      (positive = unhedged exposure)
Yield headroom           =P90_output / Output_assumed_in_debt_sizing - 1
Residual flag            =IFS(Exposure>Threshold,"RED",Exposure>0,"AMBER",TRUE,"GREEN")
```

## Hands-on: a quick P50/P90 from a yield distribution in Python

```python
import numpy as np

p50_mwh = 300_000          # illustrative annual P50 from an independent yield assessment
sigma_pct = 0.08           # illustrative one-year total uncertainty (check the yield report)
z90 = 1.2816               # standard normal quantile for 90% exceedance
p90 = p50_mwh * (1 - z90 * sigma_pct)
print(f"P90 (1-year) ≈ {p90:,.0f} MWh ({p90 / p50_mwh - 1:.1%} vs P50)")

# simulated check of the same idea
rng = np.random.default_rng(1)
sims = rng.normal(p50_mwh, p50_mwh * sigma_pct, 100_000)
print(f"Simulated P90 ≈ {np.percentile(sims, 10):,.0f} MWh")   # exceeded 90% of the time
```

This assumes a normal distribution for simplicity; independent yield assessments specify their own uncertainty, time horizon (1-year vs 10-year) and method. Use their figures in the model.

## How to measure success

- Every risk in the matrix has a mechanism, a clause reference and a quantified residual.
- No red residuals remain before lender due diligence begins.
- Debt sizing uses the lender-required production or demand case.

## Video lecture: Risk allocation and bankability

Lecture coming soon · 9 chapters · about 8 minutes. Read the full transcript below.

1. What makes lenders say yes
2. Why it matters
3. The concept: allocate to the best manager
4. Worked example one: a back-to-back check
5. Worked example two: a wind project in Sindh
6. Watch me do it: a bankability review sheet
7. P50, P90 and E&S standards
8. Common mistakes
9. Recap and try this now

## Lecture transcript

### What makes lenders say yes

A developer once told me their wind project was obviously bankable: great site, strong tariff, experienced contractor. Six months later they were still negotiating, because the lenders had found three problems nobody on the sponsor side had looked for: a cap on delay damages that was far too low, a currency mismatch, and only one year of wind data. None of those were about the wind. All of them were about risk allocation. In this lecture you'll learn what bankable really means, the core principle of risk allocation, a standard risk allocation matrix, the contract features lenders examine, how to spot back-to-back gaps, how P50 and P90 cases affect debt sizing, and why environmental and social standards can make or break a deal. By the end, you'll be able to run your own bankability review before the lenders do.

### Why it matters

Why does it matter? Because in project finance, no bankability means no debt, and usually no project. Lenders are repaid only from project cash flows, so they care intensely about how predictable those cash flows are and who carries each risk. Every gap that lenders find either becomes a condition they require the sponsor to fix, lowers the debt they'll offer, or lands back on the sponsor as support they must provide. And the later a gap is found, the more it costs: extending bid bonds, renegotiating contracts or keeping a contractor on standby. A sponsor who runs a disciplined bankability review early saves time, money and negotiating leverage.

### The concept: allocate to the best manager

Here's the core principle: allocate each risk to the party best able to manage it, at the lowest cost. Risks nobody can manage, such as some natural catastrophes, are insured, shared or priced. So construction cost overrun and delay usually sit with the EPC contractor, through a fixed-price, date-certain contract with delay liquidated damages sized to cover debt service and fixed costs. Performance shortfall sits with the EPC contractor and O&M operator through performance damages and guarantees. Demand or volume risk sits with the offtaker in contracted models, through take-or-pay agreements or availability payments, or with the SPV in merchant projects. Interest rate risk is hedged. Currency risk is handled through tariff indexation, local-currency debt or hedging. Change in law often sits with the grantor or offtaker through compensation clauses. And political risk and force majeure are shared, insured, or supported by DFI participation.

### Worked example one: a back-to-back check

Here's a simple back-to-back check, with illustrative numbers. The offtake agreement penalises the SPV sixty thousand a day if commercial operation is late. The EPC contractor's delay damages are forty thousand a day, capped at ten per cent of the contract price. So on day one of any delay, the SPV loses twenty thousand a day that nobody else pays. And once the EPC cap is reached, the SPV bears the entire sixty thousand a day, plus its own debt service and fixed costs during the delay. That's a gap. The fix might be higher delay damages, a higher cap, sponsor completion support, or delay-in-start-up insurance. The point is to find it by lining up the contracts side by side, before a lender's adviser does it for you.

### Worked example two: a wind project in Sindh

Now the realistic example from the lesson. A fictional one hundred megawatt wind project in Sindh goes to lenders. The review finds five issues. Offtake: a twenty-year power purchase agreement with a state buyer whose payment history includes delays; the fix is a payment security mechanism, sovereign support or DFI participation. Currency: the tariff is only partly indexed to dollars, while debt is partly in dollars; fix with more indexation or a local-currency tranche. EPC: delay damages capped at ten per cent of the contract price, which may be too low; review the cap against expected delay costs and add sponsor completion support. Wind resource: only one year of on-site data; get longer data, an independent yield assessment, and size debt on a conservative P90 case. And grid: connection depends on the utility; add deemed energy payments if the grid isn't ready. Outcome: lenders finance it, but at lower gearing than proposed, reflecting residual risk.

### Watch me do it: a bankability review sheet

Let me show you the review sheet I build before any lender meeting. One row per risk. Columns for who bears it, the mechanism, any cap or limit on that mechanism, the contract clause, and the residual exposure left with the SPV. The residual column is where the work happens. For delay, I compare the offtake penalty plus debt service during delay against the EPC damages and cap. For currency, I compare the indexed share of the tariff against the dollar share of debt service. For energy yield, I compare P90 production with the production the debt sizing assumes. Then conditional formatting: red if residual exposure exceeds a threshold, amber if it's covered only by sponsor support. Every row cites the clause, so when a lender asks, I can show the words, not just my summary.

### P50, P90 and E&S standards

Two more lender essentials. First, production cases. For renewables, lenders usually size debt on a conservative production estimate, often a P90 or P99 case over one or ten years, while equity may use the P50 case. P90 means production expected to be exceeded with ninety per cent probability. The same logic applies to traffic forecasts on toll roads: lenders use downside cases. Second, environmental and social standards. Many lenders, especially DFIs and banks that have adopted the Equator Principles, require compliance with international standards such as the IFC Performance Standards. Non-compliance can make a project unbankable regardless of its economics, and fixing it late, such as resettlement planning or biodiversity assessments, can delay close by many months.

### Common mistakes

The common mistakes. Assuming a contract transfers risk when its liability caps make the transfer meaningless. An offtake tenor shorter than the debt tenor, so there's no contracted revenue for the final repayments. Ignoring interface risk between the EPC, O&M and offtake contracts, where each party can point at another. Underestimating the importance of counterparty creditworthiness: a perfect contract with a weak buyer is still a weak contract. And leaving environmental and social assessments too late. A useful discipline: before any lender meeting, ask 'which risk in this matrix would I least like to explain?' That's the one to fix first.

### Recap and try this now

Let's recap. A project is bankable when lenders will finance it on acceptable terms, which comes from predictable cash flows, sensible risk allocation, creditworthy counterparties, robust contracts, enforceable security and a sound legal framework. Allocate each risk to the party best able to manage it, and check that contracts line up back to back, including their caps. Expect lenders to size debt on conservative cases like P90, and to require recognised environmental and social standards. Your try-this-now: for a project you know, build a risk allocation matrix with at least eight risks, the party bearing each, the contractual mechanism, and the residual exposure left with the SPV. Then circle the one you'd least like to explain to a lender.

## Key takeaways

- Bankability = predictable cash flows + sensible risk allocation + strong counterparties + robust, enforceable contracts.
- Allocate each risk to the party best able to manage it at the lowest cost.
- Check back-to-back alignment between EPC, O&M and offtake to avoid gaps.
- Lenders size on downside cases (e.g., P90 yield) and require environmental and social compliance.

## Try it

For a project you know, build a risk allocation matrix with at least eight risks, the party bearing each, and the contractual mechanism used.

- [Previous: Capital structure, gearing and the cost of capital](https://optimizeall.com/learn/project-finance-and-financial-modelling/capital-structure)
- [Next: Contracts and procurement basics](https://optimizeall.com/learn/project-finance-and-financial-modelling/contracts-and-procurement)
- [All lessons of Project Finance & Financial Modelling](https://optimizeall.com/learn/project-finance-and-financial-modelling)