Part II: Practice of Dividend Distribution and Modelling
Chapter 5: Dividend Policy—Multilayered Constraints and Decision Mechanism
5.1 Multilayered Constraint Structure in Dividend Determination
Dividend determination in project finance is not automatically implemented simply by satisfying DSCR criteria. Dividends can only be implemented after sequentially clearing multiple constraint conditions—a multilayered gate structure. Without understanding this structure, it is impossible to calculate distributable amounts accurately.
Dividend determination requires dual judgment on the legal source and the feasibility of implementation. Legally, dividends are paid from retained earnings on the income statement under company law. However, even if retained earnings exist, dividends cannot be implemented if the cash remaining after the waterfall cash allocation is insufficient. Therefore, the following are required: first, retained earnings must exist on the income statement; second, cash must remain after fulfilling priority obligations of debt service and reserve funding through the waterfall; third, DSCR must meet financial covenant criteria; fourth, shareholder approval must be obtained based on the shareholders’ agreement.
Table 5-1 organises the multilayered constraint structure in dividend determination and its legal effects.
Table 5-1: Multilayered Constraint Structure in Dividend Determination
| Gate | Constraint Type | Criterion | Legal Consequences if Not Met |
| Gate 1 | Legal Source | Retained Earnings > 0 | A dividend is illegal under company law |
| Gate 2 | Cash Availability | Post-Waterfall Cash > 0 | Dividend physically impossible |
| Gate 3 | DSCR Covenant | DSCR ≥ Minimum (typically 1.20) | Dividend restriction/prohibition |
| Gate 4 | LLCR Covenant | LLCR ≥ Minimum (typically 1.30) | Enhanced monitoring, cash sweep |
| Gate 5 | Cash Sweep | Based on the DSCR level | Partial/full sweep to prepayment |
| Gate 6 | Reserve Requirements | All reserves are at minimum levels | Dividend prohibited until replenished |
| Gate 7 | Shareholders Agreement | Shareholder approval obtained | Dividend decision invalid |
These gates are arranged in series; if even one cannot be passed, dividends cannot be implemented. The theoretical upper limit of distributable amount is determined as the minimum value among retained earnings on the income statement, post-waterfall cash remaining, and DSCR constraint ceiling. In practice, each constraint value is calculated independently, and the most stringent constraint determines the effective dividend cap.
5.2 DSCR-Based Dividend Restrictions: Staged Structure of Lock-up and Cash Sweep
DSCR plays a central role in dividend restrictions. Loan agreements typically stipulate Lock-up clauses and Cash Sweep clauses that restrict dividends according to DSCR levels. The Lock-up clause completely prohibits dividends when DSCR falls below a certain level. The Cash Sweep clause mandatorily applies excess cash flows to prepayment when DSCR falls below a certain level.
Table 5-2 shows the standard staged structure of dividend restrictions by DSCR level.
Table 5-2: Staged Structure of Dividend Restrictions According to DSCR Level
| DSCR Range | Dividend Treatment | Cash Sweep Rate | Economic Effect |
| ≥ 1.30 | No restrictions | 0% | Full dividend to shareholders |
| 1.25 – 1.30 | Partial sweep | 25% to prepayment | 75% available for dividend |
| 1.20 – 1.25 | Moderate sweep | 50% to prepayment | 50% available for dividend |
| 1.15 – 1.20 | Heavy sweep | 75% to prepayment | 25% available for dividend |
| 1.10 – 1.15 | Lock-up activated | 100% to prepayment | Zero dividend |
| < 1.10 | Event of Default | 100% + step-in rights | Zero dividend + lender control |
When DSCR is 1.30 or above, there are no dividend restrictions, and all cash flows after debt service can be applied to dividends. However, the legal source constraint of retained earnings still applies. When DSCR is in the 1.25-1.30 range, 25% of excess cash flows are mandatorily applied to prepayments, restricting the distributable amount to 75% of excess cash flows. In the 1.20-1.25 range, 50% of the excess is applied to prepayment. In the 1.15-1.20 range, 75% of the excess is applied to prepayment, restricting the distributable amount to 25% of the excess.
In the 1.10-1.15 range, the Lock-up clause is activated, and dividends are completely prohibited. All excess cash flows are applied to prepayment to restore DSCR. When DSCR falls below 1.10, events of default are triggered, and in addition to dividend prohibition, banks can exercise step-in rights and demand accelerated loan repayment. Through this staged structure, dividends are progressively restricted as DSCR declines, prioritising debt service.
5.3 Reserve System: Types and Priority Over Dividends
In project finance, multiple reserve accounts are established, each with funding obligations. Reserves are buffers against future cash flow fluctuations and sudden expenditures, funded with priority over dividends. Table 5-3 organises major reserve types and their functions.
Table 5-3: Major Reserve Types and Functions
| Reserve Type | Typical Amount | Purpose | Priority Rank | Impact on Dividend |
| Debt Service Reserve Account (DSRA) | 6 months debt service | Buffer for debt payment | Highest (Priority 6) | If insufficient, dividend = 0 |
| Maintenance Reserve Account (MRA) | Project-specific cycle | Major overhaul funding | High (Priority 7) | If insufficient, dividend reduced |
| Insurance Premium Reserve (IPR) | Annual premium | Insurance payment | High (Priority 7) | Default if unpaid |
| Working Capital Reserve | Initial buildup | Receivables/inventory | Medium (Priority 7) | One-time funding |
| Decommissioning Reserve | End-of-life costs | Asset removal | Medium-Low (later years) | Gradual accumulation |
The Debt Service Reserve Account is the most important reserve, standardly maintaining 6 months of next-period debt service at all times. If DSRA is insufficient, dividends are completely prohibited, and all excess cash flows are applied to replenishing DSRA. The Maintenance Reserve Account prepares for periodic equipment repairs, aligned with industry-specific maintenance cycles, such as 5-year periodic inspections for power plants and 3-year turnarounds for gas plants. If MRA is insufficient, dividends are restricted, and the distributable amount is reduced until the required amount is accumulated.
The Insurance Premium Reserve prepares for annual insurance premium payments. Non-payment of insurance premiums causes an insurance contract lapse and constitutes an event of default under loan agreements, so IPR has high priority. The Working Capital Reserve prepares for working capital needs accompanying increases in accounts receivable and inventory. During steady-state operations, working capital remains nearly constant, so after initial funding of the required amount, additional funding becomes unnecessary. The Decommissioning Reserve prepares for equipment removal costs at business termination, with funding starting in the latter half of the business period.
5.4 Calculation of Distributable Amount: Waterfall Model
Dividend implementation is subject to dual judgments: retained earnings on the income statement and post-waterfall cash remaining. Retained earnings are the legal source of dividends; under company law, dividends are illegal without their existence. Meanwhile, the waterfall shows the priority order of actual cash allocation; only cash remaining after fulfilling all priority obligations, such as debt service, reserve funding, and Cash Sweep, can be applied to dividends. The distributable amount is the smaller of retained earnings and post-waterfall cash remaining.
Table 5-4 shows the standard waterfall structure.
Table 5-4: Cash Allocation Waterfall Priority Structure and CF Classification
| Priority | Use of Funds | Cash Flow Classification | Mandatory/Discretionary |
| 1 | Operating Expenses | Operating CF outflow | Mandatory (business survival) |
| 2 | Taxes | Operating CF outflow | Mandatory (legal obligation) |
| 3 | Maintenance Capex | Investing CF outflow | Mandatory (capacity maintenance) |
| = CAFDS | Checkpoint: Debt service capacity | ||
| 4 | Interest Payment | Financing CF outflow | Mandatory (loan covenant) |
| 5 | Principal Repayment | Financing CF outflow | Mandatory (loan covenant) |
| 6 | DSRA Funding | Reserve accumulation | Mandatory (financial covenant) |
| 7 | Other Reserve Funding (MRA, IPR, etc.) | Reserve accumulation | Mandatory (financial covenant) |
| 8 | Cash Sweep (if triggered) | Financing CF (prepayment) | Conditional mandatory |
| 9 | Dividend Distribution | Financing CF outflow | Discretionary (residual) |
Priority 1 operating expenses are minimum expenditures for continuing business and have absolute priority. Priority 2 taxes are legal obligations; delinquency invites seizure or suspension of business orders. Priority 3 maintenance capital expenditure is spending to maintain existing production capacity. After these expenditures, the remaining amount becomes CAFDS, the evaluation criterion for debt service capacity.
Priority 4 interest payments and Priority 5 principal repayments are obligations under loan agreements; non-performance constitutes an event of default. Priority 6-7 reserve funding is mandated in loan agreements and forms part of financial covenants. The Priority 8 Cash Sweep is a conditional obligation that is triggered based on the DSCR level. Priority 9 dividends can be paid only from the remaining amount after fulfilling all these obligations.
A specific calculation example is shown in Table 5-5.
Table 5-5: Calculation Example of Implementable Dividend Amount (Waterfall, Year 5, Million USD)
| Item | Amount | Cumulative Remaining | Note |
| EBITDA | 40.00 | 4.000 | Starting point |
| – Tax | (9.60) | 30.40 | FCFF (Simplified) |
| – Maintenance Capex | (5.00) | 25.40 | = CAFDS |
| – Interest | (3.00) | 22.40 | |
| – Principal | (3.72) | 18.68 | |
| – Reserve Funding (DSRA + MRA) | (2.50) | 16.18 | |
| – Cash Sweep (DSCR maintenance) | (5.74) | 10.44 | Forced if DSCR<1.25 |
| = Post-Waterfall Cash | 10.44 | Available for dividend | |
| Legal Source Check: | |||
| Retained Earnings (cumulative) | Assumed 28.05+ | From net income | |
| Final Distributable Amount | 10.44 | min(Cash, Retained Earnings) |
This calculation illustrates the cash allocation process through the waterfall. CAFDS is 25.40 million USD, which becomes the evaluation criterion for debt service capacity. Sequentially deducting debt service of 6.72 million USD, reserve funding of 2.50 million USD, and Cash Sweep of 5.74 million USD from this CAFDS results in post-waterfall cash remaining of 10.44 million USD.
The distributable amount is determined by comparing the remaining cash of 10.44 million USD with retained earnings on the income statement. In this example, assuming an after-tax profit of approximately 28.05 million USD is recorded and cumulative retained earnings are sufficient, retained earnings exceed cash remaining by 10.44 million USD, so the constraining factor is cash remaining. The final distributable amount is 10.44 million USD.
If cumulative retained earnings were only 5.00 million USD, the distributable amount would be restricted to 5.00 million USD. This is a company law constraint. Also, if the cash balance is extremely insufficient and the waterfall calculation results show that even reserve funding cannot be completed, the distributable amount becomes zero. Thus, the distributable amount is determined through a three-step process: retained earnings, post-waterfall cash remaining, and DSCR constraints.
5.5 Practical Process of Dividend Policy: Decision Procedures and Frequency
Dividend implementation follows procedures based on the shareholders’ agreement. In practice, quarterly or annual dividends are common. Table 5-6 shows dividend frequency options and practical considerations.
Table 5-6: Dividend Frequency Options and Practical Considerations
| Frequency | Advantages | Disadvantages | Typical Use Case |
| Monthly | Fastest cash return to shareholders | High administrative burden | Rarely used except for small projects |
| Quarterly | Balanced: early return + manageable burden | Moderate complexity | Most common in practice |
| Semi-Annual | Moderate administrative burden | Delayed return to shareholders | Medium-sized projects |
| Annual | Minimal administrative burden, high certainty | Up to a 1-year wait for shareholders | Conservative projects, startups |
Quarterly dividends are most frequently adopted in practice because they accelerate cash flow return to shareholders while keeping administrative burden within acceptable limits. Annual dividends minimise administrative burden and have high certainty by implementing dividends after confirming cash flows throughout the year. However, shareholders must wait up to one year for dividends.
The dividend decision process typically proceeds through the following steps: First, the finance department prepares quarterly or annual financial statements and calculates CAFDS. Second, post-waterfall cash remaining is calculated based on the waterfall calculation. Third, retained earnings on the income statement are confirmed. Fourth, compliance with all financial covenants is verified. Fifth, dividend resolutions are passed at board meetings or shareholders’ meetings. Sixth, dividend implementation to shareholders and tax processing are performed. This series of processes is detailed in shareholders’ agreements and loan agreements, ensuring transparency and predictability.
In shareholders’ agreements, a maximum dividend policy or retention-focused policy is adopted as the dividend policy. The maximum dividend policy implements dividends to the maximum extent possible within financial covenant limits, prioritising cash flow returns to shareholders. Retention-focused policy discretionarily retains part of excess cash flows to prepare for future uncertainty, emphasising financial safety. In practice, these policies are selected according to project risk profiles and sponsor investment policies. The next chapter discusses methods for constructing financial models incorporating these dividend decision processes.
Chapter 6: Financial Model Construction and Operation
6.1 Role of Financial Models and Usage Phases
A Financial Model is the core tool for decision-making and contract management in project finance. Used continuously from the bidding stage through the operation period, it supports all capitalist investment decisions, lender credit reviews, and sponsor dividend plans. This chapter discusses the structure of financial models used in practice, sensitivity analysis, and third-party verification.
Financial models are calculation engines that forecast cash receipts and payments over the project’s entire period on a monthly or annual basis and calculate indicators such as IRR, DSCR, and LLCR. Their role differs by phase (Table 6-1).
Table 6-1: Major Uses of Financial Models by Phase
| Phase | Primary Purpose | Key Output | Decision Maker |
| Bidding Stage | Maximise Equity IRR with optimistic assumptions | IRR, NPV | Sponsors |
| Financing Negotiation | Demonstrate debt service capacity with conservative assumptions | DSCR, LLCR | Lenders |
| Financial Close | Establish Base Case as contractual benchmark | All indicators | All parties |
| Construction Period | Monitor cost overruns and schedule delays | CAPEX variance, revised forecasts | Sponsors, Lenders |
| Operation Period | Quarterly covenant compliance monitoring | Actual vs. Base Case variance | Lenders |
| Refinancing | Evaluate refinancing opportunities | IRR improvement, interest savings | Sponsors |
At the bidding stage, Equity IRR is maximised with optimistic assumptions, but at the financing negotiation stage, it is revised to conservative assumptions (P90 demand forecast, higher OPEX, etc.) at lenders’ request. The Base Case model agreed upon at Financial Close becomes the standard for subsequent contract management. The model is not merely a calculation tool but has the character of an agreement document regarding assumptions among all parties. Lenders monitor deviations from the Base Case (actual demand downturns, construction cost overruns, etc.) quarterly and determine whether covenant violations have occurred.
6.2 Basic Structure and Calculation Flow
Practical financial models are typically constructed in Excel format with multiple worksheets that cross-reference each other. Table 6-2 shows the standard configuration.
Table 6-2: Standard Worksheet Configuration of Financial Model
| Worksheet | Primary Content | Key Outputs | Links To |
| Assumptions | All input parameters (macro, micro, financial) | None (input only) | All sheets |
| Revenue | Sales volume, prices, escalation | Annual revenue | P/L |
| OPEX | Fixed costs, variable costs | Annual OPEX | P/L |
| CAPEX | Construction schedule, cost breakdown | Annual capex | CF, BS |
| Depreciation | Asset register, depreciation schedule | Annual depreciation | P/L, Tax |
| P/L (Income Statement) | Revenue, costs, EBITDA, net income | Net income, retained earnings | BS, CF, Tax |
| Tax | Taxable income, tax calculation, loss carryforwards | Tax payable, effective rate | P/L, CF |
| Debt Schedule | Principal, interest, balance, DSCR | Debt service, DSCR, LLCR | CF, BS |
| Cash Flow | Operating CF, investing CF, financing CF | CAFDS, FCFE | BS, Distributions |
| BS (Balance Sheet) | Assets, liabilities, equity | Cash balance, net assets | All sheets |
| Reserves | DSRA, MRA, other reserve accounts | Reserve balances | CF, Distributions |
| Distributions | Waterfall calculation, dividend determination | Distributable amount | CF, BS |
| Returns | NPV, IRR calculations | Equity IRR, Project IRR | Final output |
Regarding the relationship among the three financial statements, an accurate understanding is necessary. The Income Statement (P/L) and Balance Sheet (BS) are mandatory financial statements under accounting. The Cash Flow Statement (CF) is theoretically intended as a reference since it is derived from annual changes in BS. However, in practice, the CF sheet performs an integrated calculation of operating CF, investing CF, and financing CF. It serves as the basis for determining the distributable amount, so in model construction, it is treated as a core sheet. The role of the BS sheet is to track year-end cash balance (including breakdowns of Restricted Cash, Debt Service Reserve, etc.) and debt balance (including breakdowns of Senior Debt, Subordinated Debt, etc.), visualising changes in asset value, Net Asset (net worth) trends, and company management soundness indicators. Lenders detect BS abnormalities (excessive cash accumulation, slow debt balance reduction, etc.) even when DSCR meets the criteria, thereby uncovering model errors or contract violations.
The calculation flow begins with the Assumptions sheet, calculates profit in P/L via Revenue and OPEX, and confirms tax amounts in the Tax sheet. The Debt Schedule sheet calculates principal and interest payments and DSCR, and the Cash Flow sheet calculates CAFDS through the waterfall. The BS sheet updates the financial position at each period-end, and the Distributions sheet determines the distributable amount based on both the DSCR constraint and retained earnings. Finally, the Returns sheet calculates Equity IRR and Project IRR. The model’s calculation period covers the period from construction start to PPA or concession expiration.
Table 6-3: Standard Periods by Project Type
| Project Type | Construction Period | Debt Repayment Period | Business Period | Model Calculation Period |
| Toll Roads | 3-5 years | 20-25 years | 25-30 years (concession) | To concession end |
| Thermal Power (PPA) | 2-3 years | 15-18 years | 20-25 years (PPA) | To PPA end |
| Solar PV (PPA) | 1-2 years | 15-20 years | 20 years (FIT/PPA) | To PPA end |
| LNG Terminals | 3-4 years | 15-18 years | 20-25 years (GSA) | To GSA end |
| Oil & Gas E&P | 2-4 years | 10-12 years | 15-20 years (reserves) | To reserve depletion |
Note: Principal repayment period refers to the period excluding the grace period (no principal repayment during construction). The business period is the PPA contract or concession contract period, often continuing after principal repayment completion. The financial model’s calculation period extends to the PPA or concession expiration.
6.3 Sensitivity Analysis and Scenario Analysis
The core value of financial models lies in quantifying the impact of assumption changes on indicators. Sensitivity Analysis handles single variable changes, while Scenario Analysis handles simultaneous changes in multiple variables.
Major factors to verify in a sensitivity analysis are initial investment (CAPEX) and operating expenses (OPEX), along with revenue variation factors (demand and price). An important practical principle is that OPEX and revenue variations have a greater impact on capitalists than CAPEX variations. The structural reason is that OPEX and revenue occur annually throughout the operation period (20-25 years), whereas CAPEX is a one-time expenditure during the construction period (2-5 years), so the impact on present value accumulates. For example, if OPEX is 10% higher than forecast continuously for 20 years, its cumulative present value easily exceeds a 10% CAPEX overrun. However, if the CAPEX variation range is extremely large (e.g., construction costs exceed by 50%), this relationship may reverse. Therefore, in sensitivity analysis, it is desirable to verify variation ranges of ±10% or more for OPEX and revenue, and cover extreme cases of ±20-30% for CAPEX.
Interest rate sensitivity analysis is also important, but its impact depends on the financing ratio and the interest rate level. In a project with a 70% financing ratio, if interest rates rise by 1%, interest payments increase annually by 0.7% of the total investment. If EBITDA is around 10-15% of total investment, the impact on DSCR remains around 0.05-0.10. However, assuming rapid interest rate increases during financial crises (2-3%), the risk of DSCR falling below 1.0 materialises. Therefore, interest rate sensitivity should be verified not only for normal variations of ±0.5% but also for extreme stress cases of ±2-3%.
In scenario analysis, composite stress, where multiple adverse factors occur simultaneously, is evaluated. However, scenario settings vary greatly depending on project characteristics, location, country macroeconomic risks, and industry-specific risk factors, so general standard patterns do not exist. Lenders reference the worst actual performance in past similar projects (demand-reduction rates during the Lehman Shock, capacity utilisation declines during COVID-19, etc.) and require Severe Stress scenarios that are at least as conservative. Even if Equity IRR falls far below the cost of capital and DSCR falls below 1.0 in this scenario, financing is approved if the probability of occurrence is below 5%.
6.4 Model Verification and Third-Party Evaluation
Financial model quality determines the reliability of investment decisions. In professional verification, the validity of assumptions and the effectiveness of sensitivity analysis are emphasised more than formal error checks. Amateurs worry about formal errors, such as “whether debt balance becomes negative” or “whether formulas contain errors,” but these are minimum conditions that should naturally be cleared. Professionals question whether assumptions are overly optimistic, whether sensitivity analysis variation ranges are practically meaningful, and whether downside scenarios are truly conservative.
Table 6-4: Practical Checkpoints for Financial Model Verification
| Category | Amateur Focus | Professional Focus | Key Risk |
| Formulas | Circular references, #DIV/0 errors | Cross-sheet consistency, macro logic | Model crashes |
| Assumptions | Input completeness | Market benchmarking, conservatism level | Over-optimism |
| Tax | Calculation accuracy | Loss carryforward treatment, effective rate | Tax shield overstatement |
| Debt Schedule | Amortisation formula correctness | Cash sweep logic, covenant interaction | Covenant breach missed |
| Sensitivity | Range selection | Correlation assumptions, stress severity | Underestimating downside |
| Documentation | Formatting, presentation | Assumption traceability to source documents | Disputes |
For large projects (investment $500 million or more), lenders appoint independent technical experts (Independent Engineers, IE) and financial model auditors (Model Auditors). The IE’s role is to verify the validity of technical assumptions. Specifically, they evaluate the basis for demand forecasts, such as Traffic Study or Market Study; the construction cost estimate breakdown; the feasibility of the construction schedule; the adequacy of O&M systems; and the consistency of technical specifications with market standards. IEs typically submit verification results in two stages: Draft Report and Final Report, with lenders reflecting Final Report conclusions in financing terms. For example, if IE revises construction cost estimates upward by 10%, lenders require sponsors to increase equity.
The Model Auditor’s role is to verify the financial model’s calculation logic and the consistency of its assumptions. Specifically, they confirm the accuracy of formula links between sheets, the validity of macros and VBA code, tax law consistency of Tax calculations, contract consistency of Cash Sweep logic in Debt Schedule, validity of sensitivity analysis variation ranges, and agreement between assumptions and source documents (PPA, EPC contract, O&M contract, etc.). Model Auditors record errors discovered during verification as an Error Log and re-verify the corrected model. Lenders do not execute Financial Close without the Model Auditor’s sign-off (approval letter).
Sponsors typically bear appointment costs for IE and Model Auditor. Verification periods require 3-6 months, with verification costs of 0.1-0.3% of total investment ($0.5-1.5 million for a $500 million project). The verification process timeline follows: Draft Report submission (3 months before Financial Close), sponsor rebuttal and corrections (1 month), Final Report submission (1 month before Financial Close), and reflection in financing terms by lenders. The impact of IE and Model Auditor verification results on financing terms is significant, substantially determining the feasibility of financing execution.
Chapter 7: Industry-Specific Characteristics and International Practice
7.1 Risk Structure and Evaluation Indicators by Industry
The investment evaluation and dividend determination framework in project finance, as discussed in Chapters 1 through 6, has a structure common across industries and regions. However, in practice, industry-specific risk characteristics, country legal systems, financial market maturity, and the forms of international financial institution involvement lead to significant differences in financing terms and dividend constraints. This chapter discusses the characteristics of evaluation indicators and dividend practices in major industries (infrastructure, energy, resource development), as well as key points of international practice in emerging-country projects.
Project finance target industries are diverse but, from a risk structure perspective, can be broadly classified into four types: contracted infrastructure (toll roads, railways, airports, water/sewage), contracted energy (PPA-type power plants, LNG terminals), merchant energy (merchant power plants, renewable non-PPA), and resource development (oil & gas, mining). The nature of revenue risk in each type determines the design of financing terms and dividend constraints (Table 7-1).
Table 7-1: Risk Structure and Financing Terms by Industry
| Industry Type | Primary Revenue Risk | DSCR Standard | Financing Ratio | Repayment Period | Key Constraints |
| Contracted Infrastructure (Toll Roads) | Traffic volume uncertainty | 1.15-1.20 | 70-80% | 20-25 years | DSCR + LLCR dual control |
| Contracted Energy (PPA Power) | Offtaker credit, fuel cost | 1.20-1.25 | 70-75% | 15-18 years | DSCR-based cash sweep |
| Merchant Energy | Market price volatility | 1.35-1.50 | 50-60% | 12-15 years | Major maintenance reserves |
| Resource Development (Oil & Gas, Mining) | Commodity price + reserves | 1.40-1.60 | 60-70% | 10-12 years | Hedging mandatory |
Typically, toll roads’ contracted infrastructure earns toll revenue under concession agreements, but traffic-prediction errors are the most significant risk factor. Demand forecasts by Traffic Study are presented in three scenarios: Base Case (P50), Conservative Case (P90), and Optimistic Case (P10). However, lenders require maintaining DSCR 1.20 or above in the P90 case as a financing condition. DSCR standards for toll roads are relatively low at 1.15-1.20 because revenue seasonal variation is small and OPEX is easily predictable. The financing ratio is high at 70-80%, with principal repayment periods of 20-25 years. In dividend constraints, DSCR and LLCR dual management is standard, with Cash Sweep triggered when DSCR<1.20 or LLCR<1.30.
The typical contracted energy of PPA-type power plants is paid through PPAs with power purchasers (Offtakers): fixed capacity payments (Capacity Payment) and variable energy payments (Energy Payment). Many PPAs include Take-or-Pay clauses, so capacity payments are made regardless of actual generation volume, as long as the power plant is available. Therefore, revenue risk is limited, but fuel cost (gas, coal) variation risk and Offtaker credit risk remain. DSCR standards are 1.20-1.25; financing ratio is 70-75%; principal repayment period is 15-18 years as standard. In dividend constraints, DSCR-based Cash Sweep is core, with dividends fully applied to prepayment when DSCR<1.25.
Merchant energy merchant power plants do not have PPAs and sell electricity in wholesale electricity markets (Spot Market). Since revenue depends entirely on market prices, financing terms become significantly stricter. DSCR standards are 1.35-1.50, the financing ratio drops to 50-60%, and principal repayment periods are shortened to 12-15 years. In dividend constraints, in addition to DSCR standards, the accumulation of the Major Maintenance Reserve and the Working Capital Reserve is required. Lenders require maintaining DSCR>1.20 even in P10 price scenarios based on the standard deviation of market prices over the past 5 years.
Resource development projects (oil & gas development, mining development) face dual uncertainty of resource price volatility and reserve risk. Therefore, financing terms are most conservative, with DSCR standards of 1.40-1.60, financing ratios of 60-70%, and principal repayment periods short at 10-12 years. In dividend constraints, in addition to DSCR, hedging (futures contracts, options) compliance is mandatory. For example, when crude oil prices fall below $50/barrel, loan agreements mandate hedging more than 50% of production at $60/barrel. Hedging costs are deducted from the distributable amount, and if hedging is not executed, dividends are completely prohibited.
7.2 Project IRR Definition and Practice by Industry
The ambiguity in the Project IRR definition discussed in Chapter 3 leads to different practical solutions across industries. In contracted infrastructure, since IRR calculation periods span long concession periods (25-30 years), IRR varies significantly depending on whether the SPV’s liquidation value is included. In practice, “Definition 1” (excluding liquidation value and measuring total investment recovery rate), assuming a zero Asset Handover (asset transfer) price at concession expiration, becomes the standard. This is because many contracts for infrastructure assets, such as roads and railways, provide for a free transfer to the government upon concession expiration. On the other hand, for airports and ports, where concession extension options exist, practice is also seen of including the extension option market value as residual value.
In contracted energy, since PPA periods (20-25 years) exceed principal repayment periods (15-18 years), the question arises of whether to include dividends after principal repayment in the Project IRR calculation. In practice, “Definition 4” (SPV sale price base, reflecting market valuation) is adopted, with SPV market value at principal repayment completion (usually the present value of future CF until PPA expiration) serving as the residual value. This is because power plants are often sold to sponsors or refinanced after loan repayment.
In merchant energy and resource development, due to high revenue uncertainty, the Payback Period (investment recovery period) is emphasised more than the Project IRR. Lenders set an investment recovery period equal to 70% of the principal repayment period under the Base Case scenario (e.g., 8 years of recovery for a 12-year principal repayment) as a financing condition. This aims to mitigate the risk of a price decline in the latter half by recovering investment early.
Table 7-2: Practical Conventions of Project IRR Definition by Industry
| Industry | Standard IRR Definition | Residual Value Treatment | Calculation Period | Rationale |
| Contracted Infrastructure | Definition 1 (exclude liquidation value) | Zero (free transfer to the government) | Concession period (25-30 years) | Assets revert to the government |
| Contracted Energy (PPA) | Definition 4 (SPV sale price base) | Market value at loan repayment | To PPA end (20-25 years) | Post-repayment sale common |
| Merchant Energy | Payback Period primary | Market value (highly uncertain) | Shortened the recovery period | Early recovery preferred |
| Resource Development | Payback Period primary | Reserve depletion = zero | To reserve exhaustion (15-20 years) | Price volatility risk |
7.3 International Practice in Emerging Country Projects
Project finance in emerging countries (developing countries, emerging markets) is characterised by greater country risk (political risk, foreign exchange risk, legal system risk) than in developed countries. Therefore, the involvement of Multilateral Development Banks (MDBs) or Export Credit Agencies (ECAs) is often a prerequisite for financing execution. MDBs include the World Bank Group (IFC, MIGA), the Asian Development Bank (ADB), and the African Development Bank (AfDB), etc., and ECAs include Nippon Export and Investment Insurance (NEXI), the US Export-Import Bank (US EXIM), and UK Export Finance (UKEF), etc.
The role of MDBs and ECAs extends beyond mere financing providers to political risk guarantees, foreign exchange risk mitigation, and debt preservation enhancement through “Preferred Creditor Status.” For example, Political Risk Insurance (PRI) provided by MIGA covers four significant risks: expropriation, war/civil unrest, currency exchange restrictions, and breach of contract. In MIGA-guaranteed projects, host country governments have incentives to avoid defaulting to MIGA, so it functions as a de facto sovereign guarantee. As a result, commercial banks relax financing terms, with financing ratios rising 5-10% and interest rates declining 0.5-1.0%.
Regarding foreign exchange risk, a currency mismatch between revenue denominated in an emerging country’s currency (e.g., local-currency PPA) and US dollar-denominated financing causes DSCR to fluctuate significantly. Lenders require maintaining DSCR>1.20 even when the host country currency depreciates 20% against the US dollar. Practical countermeasures include: ① incorporating Indexation clauses (foreign exchange-linked tariff revision clauses) in PPA contracts, ② utilising Local Currency Facilities (local currency-denominated financing) provided by MDBs, ③ using hedging instruments (Currency Swap, Forward contracts). However, in emerging countries, long-term hedging markets are often underdeveloped and hedging costs are high, so ① Indexation clauses are most effective.
Table 7-3: Impact of MDB/ECA Involvement on Financing Terms
| Aspect | Without MDB/ECA | With MDB/ECA | Improvement |
| Financing Ratio | 60-65% | 70-75% | +5-10% |
| Interest Rate Spread | 3.5-4.5% | 2.5-3.5% | -1.0% |
| Principal Repayment Period | 12-15 years | 15-18 years | +3 years |
| DSCR Minimum | 1.35-1.40 | 1.25-1.30 | -0.10 |
| Political Risk Coverage | Commercial insurance (limited) | MIGA PRI (comprehensive) | Sovereign-level |
Dividend constraints in emerging-country projects are stricter than in developed countries. Typically, ① DSCR standard increase (1.30→1.35), ② LLCR standard addition (usually 1.40 or above), ③ Major Maintenance Reserve increase (double normal), ④ Debt Service Reserve Account increase from 6 months to 12 months, ⑤ submission of additional sponsor guarantees (Sponsor Support Agreement) before dividends are required. Despite these constraints, the actual commencement timing of dividends in emerging-country projects is often 3-5 years later than in developed countries.
7.4 Characteristics of Renewable Energy and New Technology Projects
Renewable energy (solar, wind, biomass) projects have zero fuel costs, so OPEX is extremely low, and the cash flow structure differs significantly from conventional power plants. Solar power OPEX is about 2-3% of electricity sales revenue (mainly panel cleaning, inverter maintenance, and insurance premiums), which is overwhelmingly lower than gas-fired power OPEX (30-40% of electricity sales revenue, including fuel costs). As a result, the renewable energy project’s EBITDA Margin is 70-80%, far exceeding that of conventional power plants (40-50%).
However, initial investment (CAPEX) varies greatly by region. For Japanese PPA-type solar power, CAPEX is $1,300-1,500/kW with a principal repayment period of 15-17 years, whereas in the Middle East/US/Australia, $500-800/kW with a repayment period of 18-22 years is standard. The main causes of this difference are land-related costs (Japan: 12% of CAPEX; overseas: 1-2%) and project scale (Japan: 10-20MW; overseas: 100MW+). The repayment period difference causes DSCR and Equity IRR to vary by 2-3% across annual principal repayment amounts. For example, for a project with a total investment of $100 million, a financing ratio of 70%, an interest rate of 7%, and a 15-year repayment period, the DSCR remains at 1.05; with a 20-year repayment period, it improves to 1.23, expanding dividend capacity.
Table 7-4: International Comparison of Solar PV PPA Projects
| Region | CAPEX ($/kW) | Debt Repayment (years) | Land Cost (% of CAPEX) | Project Scale (MW) | Equity IRR |
| Japan | 1,300-1,500 | 15-17 | 12% | 10-20 | 6-8% |
| Middle East | 500-700 | 20-22 | 1-2% | 100-200 | 10-12% |
| United States | 700-900 | 18-20 | 2-3% | 50-150 | 9-11% |
| Australia | 600-800 | 18-22 | 1-2% | 100-300 | 10-13% |
Renewable-specific risks include: ① generation volume weather dependency (annual variation in solar radiation and wind speed), ② Technology Risk (panel degradation, wind turbine failure), ③ PPA price decline trend. Lenders require a P90 case for generation volume forecasts, assuming panel output declines to 90% of the initial value after 10 years. Under dividend constraints, a Technology Reserve (technology risk reserve) is newly required, mandating accumulation for panel replacement and inverter renewal costs.
For new technology projects such as hydrogen and ammonia power generation, Technology Risk is exceptionally high, with financing ratios dropping to 50-60% and DSCR standards raised to 1.40 or above. Dividends are prohibited until actual performance confirmation for 2-3 years after commercial operation begins.
7.5 Trends in Contract Documentation in International Practice
Standardisation of contract documentation in international project finance markets varies greatly by project type. In public infrastructure projects involving MDBs (World Bank, ADB, etc.), the use of FIDIC standard contracts is made a financing condition. FIDIC provides three types: Red Book (traditional construction), Yellow Book (design-build), and Silver Book (EPC turnkey), based on a design philosophy of fairly allocating risks between the employer and the contractor. The World Bank renewed FIDIC 9 contract usage licenses in 2023, promoting standardisation.
On the other hand, in large commercial projects (oil & gas, LNG, large power plants, etc.), Bespoke contracts (individually customised contracts) drafted by investors are mainstream. While FIDIC is balanced in risk allocation, Bespoke contracts can set terms favourable to investors. In large projects ($1 billion+), through investors’ bargaining power, liquidated damages, design change procedures, force majeure definitions, etc., are set in their favour. In projects involving complex technologies or requiring location-specific legal responses, Bespoke contracts become indispensable.
In loan agreements, the adoption of the Loan Market Association (LMA) standard document is advancing. LMA formulates standard clauses for Financial Covenants, Events of Default, etc., shortening negotiation periods. In European markets, the practice is established of adding individual modifications to LMA documents as a base. However, DSCR standards, Cash Sweep trigger conditions, reserve rules, etc., require adjustments according to project-specific risks.
International arbitration clauses are also important elements. In emerging country projects, clauses entrusting dispute resolution to ICC arbitration, LCIA arbitration, and SIAC arbitration are standard, with arbitration locations being neutral sites such as London and Singapore. Arbitration awards are enforceable in over 160 countries under the New York Convention (1958), serving as the last bastion of contract performance.
Chapter 8: Conclusion
8.1 Conclusions on Three Core Issues
This thesis systematically discusses, from a capitalist’s perspective, the practical structure of investment evaluation and dividend determination in project finance. Part I (Chapters 1-4) clarified the theoretical framework of investment evaluation, and Part II (Chapters 5-7) detailed dividend determination and financial modelling practice. This chapter presents conclusions on three core issues raised throughout the thesis and discusses practical challenges capitalists face and future prospects.
First Issue: Project IRR definition ambiguity causes serious confusion in practice. As detailed in Chapter 3, at least four different definitions of Project IRR coexist, resulting in differences exceeding 3% for the same project depending on the definition. Definition 1 (excluding liquidation value, measuring total investment recovery rate) assumes zero SPV liquidation value and is calculated by comparing the initial investment and cumulative cash flows. Definition 2 (including liquidation value, reflecting market valuation of project assets) includes SPV market value at loan repayment as residual value. Definition 3 (calculated for the period until debt repayment completion, corresponding to the lender’s credit risk evaluation) ends the IRR calculation at principal repayment completion. Definition 4 (SPV sale price base, reflecting realised price in M&A market) treats the actual sale transaction price as the end-point cash flow. This diversity of definitions has different practical solutions depending on the industry. In contracted infrastructure (toll roads, railways), since assets are freely transferred to the government at concession expiration, Definition 1 becomes standard. In contracted energy (PPA-type power plants), since cases of post-loan-repayment sale to sponsors are common, Definition 4 is adopted. In merchant energy and resource development, due to high revenue uncertainty, the Payback Period (investment recovery period) is emphasised more than the Project IRR. Capitalists must explicitly agree on IRR definitions used in investment decisions and record them in contract documents.
Second Issue: The mechanism of tax shield Rd×(1-Tc) shows the basis for debt utilisation increasing project value. As analysed from three perspectives (tax perspective, cash perspective, market perspective) in Chapter 3, through tax deduction of interest paid, the interest cost actually borne by the enterprise is reduced from the nominal interest rate. From the tax perspective, since interest paid Rd is deducted from taxable income, the after-tax effective interest rate becomes Rd×(1-Tc). From a cash perspective, corporate tax payments decrease through interest payments, with the difference Rd×Tc retained as cash. From a market perspective, the present value of Tax Shield is added to the enterprise value, and WACC declines as the debt ratio increases. In project finance, since financing ratios reach 70-80%, Tax Shield effects are extremely large. For example, in a project with $70 million financing, 7% interest rate, and a 25% corporate tax rate, the annual Tax Shield reaches $1.225 million ($70 million × 7% × 25%), with a 20-year present value (7% discount rate) of $13 million. As a result, Equity IRR exceeds Project IRR by 3-5%. However, when tax loss carryforwards exist, Tax Shield during construction and early operational periods is not realised, so the IRR calculation must accurately reflect the timing of loss carryforward consumption.
Third Issue: Multilayered dividend determination constraints show that the distributable amount cannot be determined by DSCR alone. As detailed in Chapter 5, dividend determination is determined at the intersection of eight constraints. First, retained earnings (Retained Earnings) as a company law source must be positive. During construction and early operations, dividends are not possible due to accumulated deficits. Second, post-waterfall cash remaining becomes the dividend source. Only the remaining amount after operating CF, investing CF, debt service, and reserve funding becomes a dividend candidate. Third, DSCR standards (usually 1.20-1.30) must be met when DSCR<1.20, dividends are completely prohibited (Lock-up). Fourth, LLCR standards (usually 1.30-1.50) may be added. Fifth, through Cash Sweep clauses, part or all of the dividends are applied to prepayment based on the DSCR level. Sixth, reserves, such as the Major Maintenance Reserve and the Debt Service Reserve Account, must maintain minimum balances. Seventh, in emerging country projects, dividend remittances may be restricted by foreign-exchange controls. Eighth, dividend ratios are stipulated by the Shareholders’ Agreement among sponsors. These constraints are evaluated simultaneously, with the most stringent condition determining the distributable amount. In practice, even when DSCR meets standards, cases frequently occur where dividends become zero due to reserve insufficiency or Cash Sweep activation. Capitalists must simulate distributable amounts not only in Base Case scenarios but also in Moderate Stress and Severe Stress scenarios in advance and reflect them in investment decisions.
8.2 Practical Differences by Industry and Region
As discussed in Chapter 7, investment evaluation and dividend determination practices differ greatly by industry and region. By industry, contracted infrastructure (toll roads, railways) has Traffic Risk as the most significant concern, with DSCR standards relatively low at 1.15-1.25, but financing periods extending long-term to 20-25 years. Contracted energy (PPA-type power plants) focuses on Offtaker credit risk and fuel cost volatility, with DSCR standards of 1.20-1.30 and financing periods of 15-18 years. Merchant energy (merchant power plants) is dominated by market price risk, with DSCR standards raised to 1.35-1.50 and financing ratios dropping to 50-60%. Resource development (oil & gas, mining) faces dual uncertainties: resource price volatility and reserve risk. With DSCR standards at 1.40-1.60 and financing periods the shortest at 10-12 years, these are the most conservative conditions.
By region, comparing Japanese and overseas (Middle East/US/Australia) solar PV PPA projects shows prominent differences. In Japan, CAPEX is $1,300-1,500/kW, whereas overseas it is $500-800/kW, about half. The main causes of this difference are land-related costs (Japan 12%, overseas 1-2%) and project scale (Japan 10-20MW, overseas 100MW+). Furthermore, principal repayment periods differ by 3-5 years between Japan’s 15-17 years and overseas 18-22 years, with this difference having a 2-3% impact on DSCR and Equity IRR. Japanese lenders take a conservative stance of suppressing the repayment period to 75-85% against a 20-year PPA period, but 80-90% is standard overseas. This difference stems from differences in refinancing market maturity, accumulated project finance experience, and lender risk perceptions.
In emerging-country projects, country risk (political risk, foreign-exchange risk, legal-system risk) dominates all decisions. Through involvement of MDBs (World Bank, ADB, etc.) and ECAs (NEXI, US EXIM, etc.), financing ratios improve 5-10%, interest rate spreads decline 1.0%, and principal repayment periods extend 3 years. MIGA’s political risk insurance covers four major risks—expropriation, war, foreign exchange restrictions, and contract breach —and serves as a de facto sovereign guarantee.
8.3 Future Prospects and Practical Implications
Investment evaluation and dividend determination in project finance are entering a new phase due to the energy transition and the response to climate change. Renewable energy projects have extremely low OPEX ratios (2-3% of electricity sales revenue), so EBITDA Margin reaches 70-80%, but significant regional CAPEX differences directly affect financing terms. New technology projects, such as hydrogen and ammonia power generation, have extremely high Technology Risk, so financing ratios drop to 50-60% and dividend commencement is prohibited for 2-3 years after commercial operation. Capitalists must secure dual guarantees of Technology Provider technical guarantees and EPC contractor performance guarantees.
Contract documentation standardisation shows contrasting trends in MDB projects versus commercial projects. In MDB projects, the use of FIDIC standard contracts (Red Book, Yellow Book, Silver Book) is made a financing condition, emphasising transparency and fairness. On the other hand, in large commercial projects, Bespoke contracts (individually customised contracts), favourable to investors, are mainstream, setting liquidated damages, design change procedures, and force majeure definitions, etc., in their favour. In loan agreements, adoption of the LMA standard document is advancing, but DSCR standards, Cash Sweep trigger conditions, reserve rules, etc., require project-specific adjustments.
Practical implications for capitalists are summarised in three points. First, explicitly agree on the Project IRR definitions used in investment decisions and record them in contract documents. Second, when determining distributable amounts, simulate not only DSCR standards but also all eight constraint conditions, quantifying zero-dividend risk in stress scenarios. Third, understand differences in financing terms by industry and region, accurately evaluating, especially, the impact of principal repayment periods on DSCR and Equity IRR. Through this practical understanding, capitalists can appropriately control risk and return in high-uncertainty project finance investments.
Leave a Reply