[Part I] Theory of Investment Evaluation: Cash Flow, Returns, and Financial Soundness Indicators
Introduction
This thesis elucidates practical challenges in investment evaluation and dividend determination in project finance.
The first core issue is the ambiguity in the definition of Project IRR. Multiple definitions coexist—pre-tax and post-tax, before and after interest, CAFDS-based—resulting in differences exceeding 3% for the same project. An explicit definition in contracts is indispensable.
Second, this work clarifies the economic mechanism of the tax shield Rd×(1-Tc). It examines how the deductibility of interest expenses enhances returns for investors as a whole from three perspectives: cash flow, profit-and-loss calculations, and overall investor returns.
Third, it reveals the multilayered constraints on dividend determination. Not only DSCR, but seven constraints—including CAFDS generation, reserve adequacy, and Cash Sweep—determine dividends.
Part I discusses investment evaluation theory, while Part II addresses dividends and modelling practices.
Chapter 1: Basic Structure of Project Finance
1.1 Definition and Essential Differences from Corporate Finance
Project finance is a financing method that uses only the cash flows of a specific project as the source of repayment. It is characterised by limited or non-recourse structures that do not depend on the parent company’s creditworthiness, fundamentally differing from corporate finance. Table 1-1 shows the significant differences between the two.
Table 1-1: Major Differences Between Project Finance and Corporate Finance
| Aspect | Corporate Finance | Project Finance |
| Repayment Source | Entire company cash flows | Project cash flows only |
| Recourse | Full recourse to parent | Limited/non-recourse |
| Credit Assessment | Parent company creditworthiness | Project economic viability |
| Debt Ratio | ~40% | ~70% |
| Cash Flow Stability | Variable | Stabilised by long-term contracts |
| Collateral | General corporate assets | Project-specific physical assets |
In corporate finance, the entire company’s cash flows serve as the repayment source, and the parent company provides full recourse guarantees for debt. In contrast, in project finance, the project’s own economic viability is the sole criterion for lending decisions. Significant differences also exist in capital structure. While corporate finance typically has a debt ratio of around 40%, project finance commonly employs a high leverage of 70%. This difference stems from project finance’s ability to secure stable cash flows through long-term purchase agreements and provide physical assets as collateral.
1.2 SPV and the Economic Rationale of Capital Structure
In project finance, establishing a Special Purpose Vehicle (SPV) is a legal requirement. The SPV serves three primary functions. First, it provides bankruptcy remoteness. Even if the parent company fails, the SPV’s assets and liabilities are legally protected from seizure by the parent company’s creditors. This isolation allows banks to evaluate only the project’s cash flow-generating capacity, thereby separating the parent company’s credit risk.
Second, the SPV provides cash flow management functionality. All revenues are concentrated in the SPV’s dedicated account and distributed according to a waterfall structure. Table 1-2 shows the standard priority order of the waterfall structure.
Table 1-2: Cash Flow Waterfall Priority Order
| Priority | Use of Funds |
|---|---|
| 1 | Operating Expenses |
| 2 | Taxes |
| 3 | Debt Service (Interest and Principal) |
| 4 | Reserve Account Funding |
| 5 | Dividends to Shareholders |
In this structure, operating expenses, taxes, debt service, and reserve accumulation take priority, with the residual distributed to shareholders as dividends. The clarification of priorities ensures repayment certainty for creditors while shareholders explicitly bear residual risk. Third, the SPV enables risk visualisation. Project-specific risks are evaluated independently, without being mixed with the parent company’s other business risks.
The economic rationale for the standard 70% debt ratio is due to multiple factors shown in Table 1-3.
Table 1-3: Economic Rationale for High Leverage
| Factor | Mechanism | Effect |
| Tax Shield Effect | Interest expense reduces taxable income | Lowers the effective cost of capital |
| Long-term Contracts | Power purchase agreements, gas sales contracts | Stabilises revenue stream |
| Physical Collateral | Power plants, LNG terminals | Mitigates lender risk |
| Predictable Cash Flows | Contracted revenues | Enhances debt capacity |
The tax shield effect reduces the effective cost of capital as interest expenses decrease taxable income. Long-term contracts such as power purchase agreements and gas sales contracts stabilise revenue and enhance debt repayment capacity. Physical assets such as power plants and LNG terminals retain collateral value, mitigating banks’ risk. These factors interact to enable high leverage. However, excessive leverage can lead to default when completion risk or market risk increases, so appropriate levels are set for each industry.
1.3 Stakeholder Roles and Risk Allocation Principles
Project finance is an ecosystem where contracts bind multiple stakeholders. Table 1-4 shows the roles and risk burdens of major stakeholders.
Table 1-4: Roles and Risk Burdens of Major Stakeholders
| Stakeholder | Role | Risk Borne | Return/Compensation |
| Sponsors | Provide equity capital | Residual risk | All profits after debt service |
| Lenders | Provide debt | Credit risk | Fixed interest rate (capped return) |
| EPC Contractors | Construction and delivery | Completion risk, technical risk | Fixed price contract fee |
| O&M Operators | Operations and performance | Operating risk | Performance-based fees |
| Offtakers | Purchase guarantee | Market risk | Access to project output |
| Government | Permits and tax framework | Political/regulatory risk | Tax revenues |
| Insurers | Specific risk coverage | Force majeure events | Insurance premiums |
Sponsors provide capital as shareholders and make important decisions. In exchange for bearing residual risk, they enjoy all profits after debt service. Lenders provide debt and continuously monitor financial conditions. They bear credit risk but receive capped returns through fixed interest rates. EPC contractors guarantee construction and delivery schedules and bear completion risk. They assume construction cost overrun risk through fixed-price contracts and technical risk through performance guarantees.
O&M operators guarantee operations and performance and bear operating risk. They absorb unplanned outage risk through availability guarantees. Off-takers guarantee the purchase of output and bear market risk. Fixed-price clauses in power purchase agreements and take-or-pay provisions in gas sales contracts eliminate revenue uncertainty. Governments provide the legal framework for projects through permits and tax systems, affecting political and regulatory risks. Insurers underwrite specific risks, protecting other stakeholders from unpredictable events such as natural disasters and political unrest.
The fundamental principle of risk allocation is “allocating risk to those who can best control it.” EPC contractors bear completion risk. EPC contractors possess construction technology and schedule management capabilities and assume risk through fixed-price, fixed-schedule contracts. EPC contractors similarly bear technical risk through performance guarantee clauses. Operating risk is borne by O&M operators. O&M operators possess operational know-how and absorb risk through availability guarantees. Off-takers bear market risk. Offtakers possess distribution networks to end consumers and provide stable revenue to projects through long-term purchase agreements.
1.4 Risk Allocation Failure Case: Channel Tunnel
The Channel Tunnel, which opened in 1994, is a typical example of risk allocation failure. Table 1-5 compares plan versus actual results.
Table 1-5: Channel Tunnel Plan vs. Actual Results
| Item | Plan | Actual | Variance |
| Construction Cost | £4.8 billion | £10.0 billion | +108% |
| Completion Date | 1993 | 1994 | +1 year delay |
| First Year Traffic | Optimistic forecast | 50% below forecast | -50% |
| Financial Outcome | Profitable from Year 1 | Debt restructuring 1998 | Equity value destroyed 97% |
Construction costs ballooned from the initial budget of £4.8 billion to £10 billion, more than doubling. The primary cause of this inflation was the contract design, under which the banking consortium bore the completion risk. Usually, EPC contractors should assume completion risk through fixed-price contracts, but in the Channel Tunnel case, the banking consortium absorbed cost overruns. As a result, the banking consortium suffered enormous losses, and shareholder value was destroyed by 97% in the 1998 debt restructuring. This case demonstrates the danger of allocating completion risk to financial institutions without construction capabilities.
1.5 Major Contract System and Legal Functions
Project finance is a legal framework in which multiple contracts are interrelated. Table 1-6 shows the major contracts and their legal functions.
Table 1-6: Major Contract System and Legal Functions
| Contract | Parties | Legal Function | Key Provisions |
| Loan Agreement | SPV and Banks | Protects creditor rights | Financial covenants, security interests |
| EPC Contract | SPV and EPC Contractor | Transfers completion risk | Fixed price, performance guarantees, and liquidated damages |
| O&M Contract | SPV and O&M Operator | Transfers operating risk | Availability guarantee, penalty clauses |
| Power Purchase Agreement (PPA) / Gas Sales Agreement | SPV and Offtaker | Guarantees revenue | Take-or-pay, minimum purchase volume |
| Shareholders Agreement | Among Sponsors | Establishes governance | Voting rights, dividend policy, and exit strategy |
| Direct Agreement | Banks and EPC/O&M | Step-in rights for banks | Substitution rights upon default |
The loan agreement is concluded between the SPV and banks, protecting creditors through financial covenants and security interests. Representative examples of financial covenants are shown in Table 1-7.
Table 1-7: Standard Financial Covenants
| Covenant Type | Typical Threshold | Consequence of Breach |
| Minimum DSCR | ≥1.20 | Dividend restriction |
| Lock-up DSCR | ≥1.15 | Dividend prohibition |
| Event of Default DSCR | ≥1.10 | Loan acceleration rights |
| Maximum Debt/Equity Ratio | ≤70/30 | Mandatory equity injection |
| Minimum Reserve Accounts | 6 months debt service | Cash sweep activation |
If these covenants are violated, events of default occur, and banks can exercise their step-in rights. EPC contracts are concluded between the SPV and EPC contractors, transferring completion risk through fixed prices, performance guarantees, and liquidated damages clauses. O&M contracts are concluded between the SPV and O&M operators, transferring operating risk through availability guarantees and penalty clauses. Power purchase agreements or gas sales agreements are concluded between the SPV and offtakers, guaranteeing revenue through take-or-pay clauses and minimum purchase quantities.
Shareholders’ agreements are concluded among sponsors to establish governance by stipulating voting rights, dividend policy, and exit strategies. Direct agreements are concluded between banks and EPC contractors/O&M operators, defining banks’ step-in rights and contractors’ substitution rights in the event of the SPV’s default. Through these rights, banks ensure project continuity and enhance debt recovery potential. The legal stability of project finance is secured through the interrelation of these contract groups. The next chapter discusses the definitions and calculation methods for cash flows generated under these contracts.
Chapter 2: Cash Flow Calculation and Analysis
2.1 Basic Cash Flow Concepts and Definitional Ambiguity
Cash flow concepts in project finance coexist in various definitions, causing practical confusion. The most basic concept is EBITDA. EBITDA means Earnings Before Interest, Taxes, Depreciation, and Amortisation, indicating cash generation capability from operating activities¹. Since depreciation is an accounting expense that does not involve a cash outflow, adding it back allows us to grasp cash-based profits.
FCFF is the abbreviation for Free Cash Flow to Firm, referring to cash flows attributable to the entire enterprise. The complete definition deducts taxes, capital expenditures, and increases in working capital from EBITDA. However, during steady-state operations, when capital expenditure approximately equals depreciation and working capital increases are close to zero, a simplified version deducting only taxes from EBITDA is also used. This simplified version is frequently used in practice, but cannot be applied during construction or expansion periods.
FCFE stands for Free Cash Flow to Equity, which refers to cash flows attributable to shareholders. It is calculated by deducting principal and interest payments from FCFF and adding new borrowings. It indicates financial covenants further constrain the theoretical upper limit of distributable dividends, but not actual dividend amounts. CAFDS stands for Cash Available for Debt Service, referring to cash flows available for debt service. The difference from FCFF is that only maintenance capital expenditures are deducted, excluding expansion capital expenditures. Banks prioritise this indicator and use it to calculate the debt service coverage ratio.
Table 2-1 organises the definitions and interrelationships of these cash flow concepts.
Table 2-1: Cash Flow Concept Definitions and Calculation Formulas
| Cash Flow Metric | Calculation Formula | Purpose |
| EBITDA | Operating Income + Depreciation & Amortization | Operating cash generation capacity |
| FCFF (Full Version) | EBITDA – Taxes – Total Capex – ΔWorking Capital | Enterprise value evaluation |
| FCFF (Simplified) | EBITDA – Taxes | Steady-state enterprise valuation |
| CAFDS | EBITDA – Taxes – Maintenance Capex | Debt service capacity |
| FCFE | FCFF – Principal & Interest + New Borrowings | Theoretical maximum dividend |
This diversity of definitions creates ambiguity in project finance contracts. Particularly when calculating Project IRR, results vary significantly depending on the cash flow definition used. This issue is detailed in Chapter 3.
Footnote
¹ Strictly speaking, EBITDA is a profit indicator on the income statement and differs from the cash flow statement. However, in project finance practice, under the assumption that during steady-state operations capital expenditure approximately equals depreciation and that working capital increases are close to zero, treating EBITDA as a proxy for operating cash generation capacity has become an international industry standard. This practice can be confirmed in the World Bank PPP Guide (2017), Moody’s rating methodology (2022), Yescombe’s “Principles of Project Finance” (2014), and others. This paper also follows this industry practice, treating EBITDA as a “cash evaluation indicator.”
2.2 Free Cash Flow Calculation Structure
Free cash flow is calculated from the income statement and adjusted for non-cash items and investments. Table 2-2 shows the standard calculation flow.
Table 2-2: Free Cash Flow (FCFF) Calculation Flow
| Item | Adjustment | Reason |
| EBITDA | Starting point | Operating cash generation |
| + Depreciation & Amortization | Add back | Non-cash expense |
| – Taxes | Deduct | Actual cash outflow |
| – Capital Expenditure | Deduct | Cash outflow for asset acquisition |
| – Δ Working Capital | Deduct | Cash tied up in receivables/inventory |
| = FCFF | Result | Cash available to all investors |
Adding back depreciation is necessary because, although it is an accounting expense, it does not involve a cash outflow, allowing an accurate grasp of cash generation capacity. Tax deduction is natural, as it is an actual cash outflow. However, attention must be paid to the tax shield effect, where interest expense deductibility reduces taxes. This effect is detailed in Chapter 3.
Capital expenditure deduction is necessary as it is a cash outflow for asset acquisition. However, it is necessary to distinguish between maintenance capital expenditure and expansion capital expenditure. Maintenance capital expenditure is spending to maintain existing production capacity and, in the long term, approximately equals depreciation. Expansion capital expenditure is spending to enhance production capacity and bring future cash flow increases. In the CAFDS calculation, only maintenance capital expenditure is deducted, excluding expansion capital expenditure.
Working capital increase deduction is necessary because increases in accounts receivable and inventory immobilise cash. Working capital is current assets minus current liabilities, with accounts receivable, inventory, and accounts payable as major components. As sales increase, accounts receivable and inventory increase, increasing working capital. This increase means cash outflow and is therefore deducted from cash flow. During steady-state operations, sales stabilise and working capital increases approach zero, making the simplified calculation formula applicable.
2.3 Peculiarity of Cash Available for Debt Service (CAFDS)
CAFDS is a special cash flow concept used by banks to evaluate debt service capacity. The major difference from FCFF lies in the treatment of capital expenditure. Table 2-3 shows the CAFDS calculation structure.
Table 2-3: CAFDS (Cash Available for Debt Service) Calculation Structure
| Item | Treatment | Rationale |
| EBITDA | Starting point | Operating cash generation |
| – Taxes | Deduct | Actual cash outflow |
| – Maintenance Capex | Deduct | Required to maintain current capacity |
| – Expansion Capex | NOT deducted | Discretionary, subordinate to debt service |
| = CAFDS | Result | Cash available for debt service |
The reason CAFDS does not deduct expansion capital expenditure is that expansion investment is considered discretionary spending that does not take priority over debt service. Banks’ concern is the level of cash flow that can reliably service existing debt, with future business expansion being a secondary concern. Therefore, only maintenance capital expenditure is deducted to evaluate the debt service capacity, assuming maintenance of current production capacity.
In practice, distinguishing between maintenance and expansion capital expenditure can be difficult. In such cases, depreciation is often used as a proxy variable for maintenance capital expenditure. This is based on the assumption that in the long term, maintenance capital expenditure approximately equals depreciation. However, short-term deviations occur, so using multi-year average values is recommended.
2.4 Tax Calculation and Preliminary Consideration of Tax Shield
In cash flow calculations, taxes are an important deduction item, but their calculation must consider the effect of interest expense deductibility. Table 2-4 shows the basic structure of tax calculation.
Table 2-4: Basic Structure of Tax Calculation
| Item | Amount (Million USD) |
| EBITDA | 10.00 |
| – Depreciation | (2.00) |
| = EBIT | 8.00 |
| – Interest Expense | (3.00) |
| = Taxable Income | 5.00 |
| × Tax Rate | 30% |
| = Tax Amount | 1.50 |
An important point in this calculation is that interest paid is deducted from taxable income. If there were no interest expense, taxable income would be 8.00 million USD, and the tax amount would be 2.40 million USD. Through interest expense deductibility, the tax amount decreases to 1.50 million USD, resulting in tax savings of 0.90 million USD. These tax savings equal the 3.00 million USD in interest expense multiplied by the 30% tax rate.
This effect, called the tax shield, is the theoretical basis for using Rd×(1-Tc) as the debt cost in calculating the weighted average cost of capital. However, for this effect to function, the project must generate sufficient taxable income. During construction or loss-making periods, even after deducting interest expense, taxable income becomes negative, and the tax shield does not function. In this case, it is carried forward as a loss carryforward to future periods, but its present value is discounted. The detailed mechanism and economic significance of the tax shield are discussed in Chapter 3.
2.5 Case Study: Power Plant Project Cash Flow Calculation
To concretise theoretical considerations, cash flow calculation is illustrated using a hypothetical power plant project. Table 2-5 shows the project’s basic assumptions.
Table 2-5: Power Plant Project Basic Assumptions
| Item | Value |
| Annual Revenue | 60.00 million USD |
| Variable Costs | 15.00 million USD |
| Fixed Costs | 5.00 million USD |
| Depreciation | 5.00 million USD |
| Interest Expense | 3.00 million USD |
| Tax Rate | 30% |
| Maintenance Capex | 5.00 million USD |
| Debt Service (Principal) | 3.72 million USD |
Based on these assumptions, cash flows for the steady-state operation period (Year 5) are calculated. Table 2-6 shows the detailed calculation process.
Table 2-6: Year 5 Cash Flow Calculation (Million USD)
| Item | Amount | Note |
| Revenue | 60.00 | |
| – Variable Costs | (15.00) | |
| – Fixed Costs | (5.00) | |
| = EBITDA | 40.00 | |
| – Depreciation | (5.00) | |
| = EBIT | 35.00 | |
| – Interest | (3.00) | |
| = EBT | 32.00 | |
| – Tax (30%) | (9.60) | |
| = Net Income | 22.40 | |
| Cash Flow Calculation: | ||
| EBITDA | 40.00 | |
| – Tax | (9.60) | |
| = FCFF (Simplified) | 30.40 | Enterprise value evaluation |
| – Maintenance Capex | (5.00) | |
| = CAFDS | 25.40 | Debt service capacity |
| DSCR Calculation: | ||
| CAFDS | 25.40 | |
| ÷ Debt Service (Interest + Principal) | ÷ (3.00 + 3.72) | |
| = DSCR | 3.78x | Well above minimum 1.20x |
From this calculation, Year 5 DSCR is 3.78 times, an extremely high level. Since the standard minimum DSCR requirement for power plant projects is 1.20 times, this level indicates sufficient margin in debt service capacity. However, this high level is a calculation result for the steady-state operation period; DSCR declines during construction and initial operation periods.
The simplified FCFF is 30.40 million USD, discounted by the weighted average cost of capital—the weighted average of equity and debt costs—to calculate enterprise value. CAFDS is 25.40 million USD, the result after deducting maintenance capital expenditure of 5.00 million USD. This amount serves as the source of debt service and dividends.
2.6 Usage Differentiation of Cash Flow Concepts and Practical Considerations
In project finance, it is necessary to select the appropriate cash flow concept according to purpose. Table 2-7 organises major evaluation indicators and their corresponding cash flow concepts.
Table 2-7: Correspondence Between Evaluation Purpose and Cash Flow Concepts
| Evaluation Purpose | Cash Flow Metric | Discount Rate | Key Users |
| Enterprise Value | FCFF | WACC | Sponsors, Investors |
| Equity Value | FCFE | Cost of Equity | Shareholders |
| Debt Service Capacity | CAFDS | N/A (ratio to debt service) | Lenders |
| Dividend Capacity | FCFE – Reserves | N/A | Shareholders |
For enterprise value evaluation, FCFF is used and discounted by the weighted average cost of capital. This method is neutral with respect to capital structure and can evaluate the project’s economic viability. For shareholder value evaluation, FCFE is used and discounted by the cost of equity. This method can directly evaluate investment value for shareholders. For debt service capacity evaluation, CAFDS is used and divided by debt service to calculate DSCR. This method is the indicator that banks prioritise most in lending decisions.
For dividend capacity evaluation, reserve funding is further deducted from FCFE. The residual after deducting debt service reserves, maintenance reserves, and other restrictive reserves becomes the theoretical dividend capacity. However, actual dividend amounts are further constrained by financial covenants, shareholders’ agreements, and corporate law. This complex constraint structure is detailed in Chapter 5.
As an important practical consideration, the principle of conservatism in cash flow forecasting applies. Banks emphasise downside risk, so revenue forecasts must be conservative and cost forecasts must include margins. Particularly for capacity utilisation rates, sales prices, and variable costs, a reasonable basis grounded in historical performance and market trends is required. Additionally, quantitatively demonstrating the impact of major assumption changes on cash flows through sensitivity analysis and scenario analysis is standard practice. The next chapter discusses IRR and WACC, investment evaluation indicators using these cash flows, in detail.
Chapter 3: IRR and WACC: Theory and Practice of Investment Evaluation
3.1 Ambiguity in Project IRR Definition and Practical Problems
Project IRR is one of the most important evaluation indicators in project finance. However, this indicator has the fundamental problem of definitional ambiguity. In practice, the definition using pre-tax, pre-interest cash flows is most common, but when contracts lack an explicit definition, interpretation disputes arise between parties.
Table 3-1 organises the major definitions of Project IRR and their characteristics.
Table 3-1: Four Definitions of Project IRR and Their Characteristics
| Definition | Cash Flow Base | Characteristics | Advantages | Disadvantages |
| Definition 1 | Pre-tax, Pre-interest FCFF | Most commonly used | Capital structure neutral | Ignores the tax shield effect |
| Definition 2 | Post-tax, Pre-interest FCFF | Theoretically rigorous | Reflects tax shield partially | Computation complexity |
| Definition 3 | Comprehensive CF (undefined) | Ambiguous | None | Causes disputes |
| Definition 4 | CAFDS-based | Lender perspective | Debt service certainty | Lacks shareholder perspective |
Definition 1, using pre-tax, pre-interest FCFF, is most frequently used in practice. The advantage of this definition is that it is neutral to capital structure. Since taxes and interest are not considered, changes in debt ratios do not affect the IRR, allowing the project’s inherent economic viability to be evaluated. It is useful when sponsors compare multiple investment opportunities, as it will enable them to compare pure business viability, excluding differences in capital structure. However, this definition completely ignores tax effects and cannot reflect the tax shield effect.
Definition 2, using post-tax, pre-interest FCFF, is the standard method for enterprise value evaluation. Since it considers taxes, it can partially reflect the tax shield effect. However, since interest paid is deducted from taxable income, the capital structure affects the tax calculation. This method is theoretically the most rigorous but faces the practical difficulty of high computational complexity.
Definition 3, using comprehensive cash flows, is an ambiguous definition that arises when there is no clear definition in contracts. In this case, the parties’ interpretations differ, leading to disputes. Definition 4 is the CAFDS-based IRR, used by banks to evaluate debt service capacity. This definition emphasises debt service certainty but lacks a shareholder perspective, making it unsuitable for overall project economic evaluation.
Table 3-2 shows IRR differences when applying the four definitions to the same project.
Table 3-2: Project IRR Differences by Definition (Power Plant Project Example)
| Definition | Project IRR | Difference from Baseline |
| Definition 1: Pre-tax, Pre-interest | 12.5% | Baseline |
| Definition 2: Post-tax, Pre-interest | 10.8% | -1.7% |
| Definition 3: Ambiguous Total CF | 10.0% – 13.0% | Variable (dispute risk) |
| Definition 4: CAFDS-based | 9.3% | -3.2% |
As this table shows, even for the same project, IRR varies from 9.3% to 12.5% depending on the definition—a difference of 3.2 percentage points. This difference has a decisive impact on investment decisions. If the sponsor’s required IRR is 11%, the investment would be approved under Definitions 1 and 2 but might be rejected under Definition 4. Therefore, explicitly defining Project IRR in contracts is a legal necessity.
As a practical countermeasure, calculating IRR using multiple definitions and specifying each in contracts is recommended. For example, setting various criteria in the form “Pre-tax Project IRR shall be maintained at 12.5% or above, Post-tax Project IRR at 10.8% or above.” This method can avoid disputes arising from definitional ambiguity. Additionally, IRR calculation assumptions—particularly the valuation date, cash flow recognition timing, and rounding methods—must also be specified in contracts.
3.2 Theoretical Foundation of Weighted Average Cost of Capital (WACC)
The weighted average cost of capital is an indicator that represents the weighted average cost of capital procurement and serves as the discount rate standard for investment decisions. The basic WACC formula is as follows:
WACC = (E/V) × Re + (D/V) × Rd × (1-Tc)
Where E is the market value of equity, D is the market value of debt, V=E+D is the enterprise value, Re is the cost of equity, Rd is the cost of debt, and Tc is the corporate tax rate. To understand the economic meaning of this formula, each component is examined in detail.
Table 3-3 organises the components of WACC and their economic meanings.
Table 3-3: WACC Components and Economic Meanings
| Component | Symbol | Typical Value | Economic Meaning |
| Equity Ratio | E/V | 30% | Proportion of equity financing |
| Debt Ratio | D/V | 70% | Proportion of debt financing |
| Cost of Equity | Re | 12-15% | Sponsor’s required return rate |
| Cost of Debt | Rd | 4-5% | Bank lending rate |
| Tax Rate | Tc | 30% | Corporate tax rate |
| Tax Shield Factor | (1-Tc) | 70% | Interest expense deductibility effect |
| After-tax Debt Cost | Rd×(1-Tc) | 2.8-3.5% | Effective debt cost to investors |
Equity ratio and debt ratio reflect the project’s capital structure. In project finance, a 70% debt ratio is standard, but it varies by industry and risk level. Cost of equity is the return rate sponsors require on investment. This value is determined by shareholders’ risk tolerance, alternative investment opportunities, and project-specific risks. The cost of debt is the interest rate banks require for lending. This value is the risk-free rate plus the project’s credit spread, generally in the range of risk-free rate + 2-3%.
The corporate tax rate is the sum of national and local taxes, with approximately 30% being standard in Japan. The tax shield coefficient (1-Tc) reflects the tax savings effect from interest expense deductibility. Through this coefficient, the effective debt cost is reduced to Rd×(1-Tc). Why (1-Tc) is not multiplied by the cost of equity but only by the cost of debt—the economic mechanism of this asymmetry becomes the subject of the next section.
3.3 Economic Mechanism of Tax Shield Rd×(1-Tc)
The tax shield is one of the most important financial effects in project finance. Through the deductibility of interest expenses, taxable income decreases, and taxes are reduced. These tax savings enhance returns for investors as a whole. However, why is (1-Tc) multiplied only by debt cost and not by the cost of equity? This asymmetry is clarified from three perspectives.
The first perspective is the cash flow perspective. Table 3-4 shows the cash flow effect of interest expenses.
Table 3-4: Cash Flow Effect of Interest Expenses (Million USD)
| Item | Without Debt | With Debt | Difference |
| EBIT | 10.00 | 10.00 | 0 |
| – Interest Expense | 0 | (3.00) | (3.00) |
| = Taxable Income | 10.00 | 7.00 | (3.00) |
| – Tax (30%) | (3.00) | (2.10) | +0.90 |
| = Net Income | 7.00 | 4.90 | (2.10) |
| + Interest Paid to Lenders | 0 | 3.00 | +3.00 |
| = Total to Investors | 7.00 | 7.90 | +0.90 |
| Effective Cost Analysis: | |||
| Nominal Interest | – | 3.00 | |
| Tax Savings | – | 0.90 | |
| Net Cash Outflow | – | 2.10 | = 3.00 × (1-0.30) |
With debt, 300 million USD of interest is paid, but taxable income decreases by 300 million USD, so taxes decrease by 90 million USD. The effective cash outflow, combining interest payment of 300 million USD and tax reduction of 90 million USD, is 210 million USD. This equals the nominal interest of 300 million USD multiplied by (1-Tc). Therefore, from the perspective of investors as a whole, the effective debt cost becomes Rd×(1-Tc).
The second perspective is the profit-and-loss calculation perspective. Table 3-5 shows the mechanism for interest deductibility.
Table 3-5: Interest Deductibility Mechanism (Million USD)
| Item | Without Interest Deduction | With Interest Deduction | Tax Shield Effect |
| EBIT | 10.00 | 10.00 | – |
| Interest Deduction | Not allowed | (3.00) | -3.00 taxable income |
| Taxable Income | 10.00 | 7.00 | -3.00 |
| Tax (30%) | 3.00 | 2.10 | -0.90 |
| Tax Savings | – | 0.90 | = 300 × 30% |
Through the deduction of 3.00 million USD of interest expense from taxable income, the tax amount decreases from 2.40 million USD to 1.50 million USD—a reduction of 0.90 million USD. This tax savings of 0.90 million USD equals the 3.00 million USD in interest expense multiplied by the 30% tax rate. These tax savings can be interpreted as a de facto subsidy from the government to investors as a whole. By not collecting 0.90 million USD of taxes that should be collected, that amount remains in investors’ hands.
The third perspective is the overall investor perspective. Table 3-6 shows cash flow distribution to investors as a whole.
Table 3-6: Cash Flow Distribution to Overall Investors (Million USD)
| Recipient | Without Debt | With Debt | Difference |
| To Lenders (Interest) | 0 | 3.00 | +3.00 |
| To Shareholders (Net Income) | 7.00 | 4.90 | -2.10 |
| To Government (Taxes) | 3.00 | 2.10 | -0.90 |
| Total to Investors | 7.00 | 7.90 | +0.90 |
With debt, the total amount investors receive is 7.90 million USD, which is 0.90 million USD more than the 7.00 million USD without debt. This increase equals the 0.90 million USD decrease in payment to the government. The tax reduction directly increases profits for investors as a whole. Since this 0.90 million USD is the interest on 3.00 million USD multiplied by a 30% tax rate, the tax shield effect is quantitatively confirmed.
So why don’t dividends have a similar effect? Table 3-7 shows the tax asymmetry between dividends and interest.
Table 3-7: Tax Asymmetry Between Dividends and Interest
| Item | Interest | Dividends | Key Difference |
| Deductibility | Tax deductible | NOT tax deductible | Tax law provision |
| Effect on Taxable Income | Reduces taxable income | No effect | Interest only |
| Tax Savings Generated | Yes: Interest × Tc | No | Fundamental asymmetry |
| Rationale | Business expense | Profit distribution | Conceptual basis |
Dividends are paid from after-tax profits and therefore do not affect the calculation of taxable income. Thus, increasing dividends does not reduce taxes, and there is no tax shield effect. This asymmetry originates from tax law provisions. Interest is considered a necessary expense for corporate business activities and is deductible, but dividends are considered profit distributions and are not deductible. This difference is why (1-Tc) is multiplied only by debt cost in the WACC calculation formula.
3.4 Situations Where the Tax Shield Does Not Function
The tax shield does not always function. Table 3-8 organises the major situations in which the tax shield does not function.
Table 3-8: Situations Where the Tax Shield Does Not Function
| Situation | Mechanism | Effect on Tax Shield | Mitigation |
| Loss-making Period | Negative taxable income | No current tax benefit | Loss carryforward (discounted) |
| Construction Period | No revenue | Capitalize interest | Deferred to depreciation period |
| Loss Carryforward Limits | Accumulated losses exceed limits | Partial non-deductibility | Proper tax planning |
| Tax Haven / Tax-exempt | No corporate tax | No tax benefit | WACC uses Rd not Rd×(1-Tc) |
During loss-making periods, taxable income is negative, so even after deducting interest expense, no tax is incurred. Therefore, no tax savings arise from the tax shield. However, many tax systems have loss carryforward provisions that allow current-period losses to offset future-period profits. In this case, the tax shield is deferred to the future, but the present value decreases due to the time value of money.
During the construction period, revenue is not yet generated, inevitably resulting in losses. Interest expenses during this period are capitalised and included in the acquisition cost of fixed assets. Capitalised interest is expensed in future periods through depreciation, so the tax shield effect is also deferred. Loss carryforwards may have deduction limits, and if accumulated losses become enormous, complete deduction may not be possible.
In tax havens or tax-exempt businesses, corporate tax is not levied in the first place, so there is no tax shield effect. In this case, the (1-Tc) term is deleted from the WACC calculation formula, and debt cost becomes Rd itself. In practice, the present value of tax shields over the project’s lifecycle is calculated, and an effective tax rate is derived.
3.5 Calculation Methods for Cost of Equity (Re)
Cost of equity is the return rate sponsors require on investment and is an important component of WACC. However, in project finance, since there is no stock market, the standard CAPM (Capital Asset Pricing Model) cannot be applied. Table 3-9 compares major calculation methods for the cost of equity.
Table 3-9: Cost of Equity (Re) Calculation Method Comparison
| Method | Formula / Approach | Advantages | Disadvantages | Typical Range |
| CAPM | Rf + β(Rm – Rf) | Theoretically sound | Beta unavailable for SPVs | N/A |
| Required IRR Method | Use sponsor’s hurdle rate | Aligns with decision-making | Subjective | 12-15% |
| Benchmark Method | Reference comparable projects | Market-based | Perfect comparables rare | Varies |
| Build-up Method | Rf + Risk Premiums | Explicit risk factors | Subjective premium estimates | 10-18% |
| Dividend Discount Model | Reverse calculation from dividends | Equity-holder perspective | Irregular dividends in PF | N/A |
CAPM adds the beta coefficient multiplied by the market risk premium to the risk-free rate. It is theoretically the most sophisticated method, but it has the fatal problem in project finance of being unable to estimate the beta coefficient. For listed companies, beta can be calculated from market data, but for SPVs, comparable projects are often unlisted. Therefore, the CAPM’s applicability in project finance is low.
The required IRR method uses the required IRR set by sponsors as their investment decision criterion directly as the cost of equity. It is the most frequently used method in practice and is consistent with sponsor investment decision-making. However, this method lacks a theoretical basis and depends on sponsors’ subjective judgment. Typically, the required IRR is in the 12-15% range.
The benchmark method references the actual IRR of comparable projects. When projects within the same industry, region, and risk profile exist, their actual IRRs serve as useful reference points. However, finding perfectly comparable projects is difficult, and adjustments are necessary. Table 3-10 shows the standard cost of equity levels by industry.
Table 3-10: Cost of Equity Levels by Industry
| Industry | Typical Cost of Equity | Key Risk Factors |
| Toll Roads (Government Guaranteed) | 10-12% | Low traffic risk, political support |
| Renewable Energy (FIT) | 8-12% | Fixed tariff, low volume risk |
| Thermal Power (PPA) | 12-14% | Fuel price risk, technology risk |
| Oil & Gas E&P | 15-20% | Commodity price, reserve uncertainty |
| Mining | 18-25% | High geological, political risk |
The build-up method adds various risk premiums to the risk-free rate. Country risk premium, industry risk premium, and project-specific risk premium are individually evaluated and summed. This method has the advantage of explicitly decomposing risk factors, but has the limitation that the quantification of each premium becomes subjective.
The dividend discount model reverse-calculates the cost of equity by assuming that the present value of dividends discounted at the cost of equity equals the equity value. However, in project finance, dividends are irregular, and growth rate estimation is also difficult, so applicability is low.
In practice, using multiple methods in combination and mutually verifying validity is recommended. For example, using the sponsor-required IRR as a baseline, comparing it with market levels using the benchmark method, and confirming risk factors using the build-up method. This triangulation approach can reduce arbitrariness.
3.6 WACC Calculation Example and Sensitivity Analysis
To concretise theoretical considerations, the WACC for a power plant project is calculated. Table 3-11 shows the calculation assumptions.
Table 3-11: WACC Calculation Assumptions
| Parameter | Value |
| Equity Ratio (E/V) | 30% |
| Debt Ratio (D/V) | 70% |
| Cost of Equity (Re) | 13.0% |
| Cost of Debt (Rd) | 4.0% |
| Corporate Tax Rate (Tc) | 30% |
Based on these assumptions, WACC is calculated:
WACC = (E/V) × Re + (D/V) × Rd × (1-Tc)
WACC = 30% × 13.0% + 70% × 4.0% × (1-30%)
WACC = 3.9% + 1.96%
WACC = 5.86%
This calculation result shows that the overall project cost of capital is 5.86%. If Project IRR exceeds this level, the project is judged economically sound. The equity cost contribution is 3.9%, and the debt cost contribution is 1.96%. Through the tax shield effect, the debt cost contribution is reduced from the nominal value of 2.8% to 1.96%—a reduction of 0.84%.
Table 3-12 shows sensitivity analysis results on the impact of major parameter changes on WACC.
Table 3-12: WACC Sensitivity Analysis
| Parameter Change | Base Case | Scenario | New WACC | Change |
| Base Case | Re=13%, Rd=4%, D/V=70%, Tc=30% | – | 5.86% | – |
| Cost of Equity +2% | 13% | 15% | 6.46% | +0.60% |
| Cost of Debt +1% | 4% | 5% | 6.35% | +0.49% |
| Debt Ratio +10% | 70% | 80% | 5.76% | -0.10% |
| Tax Rate +5% | 30% | 35% | 5.78% | -0.08% |
When the cost of equity rises from 13% to 15%—a 2% increase—WACC increases from 5.86% to 6.46%, a 0.60% increase. When the cost of debt rises from 4% to 5%—a 1% increase—WACC increases to 6.35%, a 0.49% increase. Increased debt ratio slightly lowers WACC, but the effect is limited. An increased corporate tax rate strengthens the tax shield effect and lowers WACC, but the effect is also limited. From this analysis, WACC is most sensitive to the absolute levels of the cost of equity and the cost of debt, with the impacts of capital structure and the tax rate being relatively small.
3.7 Relationship Between IRR and WACC: Investment Decision Criteria
The relationship between IRR and WACC constitutes the basic principle of investment decisions. In the net present value method, future cash flows are discounted by WACC to calculate NPV. When a project’s IRR exceeds WACC, NPV discounted at WACC is positive, and the project is judged economically sound. When IRR is below WACC, NPV discounted at WACC is negative, and the project is judged economically inappropriate. Table 3-13 organises the relationship between IRR and NPV when the WACC is used as the discount rate.
Table 3-13: Investment Decision Criteria by IRR and WACC (NPV Discount Rate = WACC)
| Condition | NPV (Discounted at WACC) | Investment Decision |
| IRR > WACC | NPV > 0 | Accept project |
| IRR = WACC | NPV = 0 | Indifferent |
| IRR < WACC | NPV < 0 | Reject project |
This relationship is a mathematical consequence derived from the definition of IRR. IRR is the discount rate that sets NPV to zero. Therefore, when the discount rate is set to WACC, if IRR is higher than WACC, NPV is positive; if IRR is lower than WACC, NPV is negative. Through this relationship, the IRR method and NPV method produce consistent investment decisions.
In practice, the difference between IRR and WACC—the spread—is an indicator of the project’s economic attractiveness. The larger the spread, the more attractive the project. Table 3-14 shows the relationship between spread levels and investment decisions.
Table 3-14: Spread Levels and Investment Decisions
| Spread (IRR – WACC) | Interpretation | Investment Decision |
| > 5% | Highly attractive | Strong acceptance |
| 3-5% | Attractive | Standard acceptance |
| 1-3% | Marginal | Careful review needed |
| 0-1% | Barely acceptable | High sensitivity to assumptions |
| < 0% | Economically unsound | Reject |
In project finance, a spread of 3-5% is typically required. This level serves as a buffer, considering forecast uncertainty and model risk. When the spread is less than 1%, NPV can turn negative with slight changes in assumptions, requiring careful examination. The next chapter discusses DSCR and LLCR, indicators for evaluating debt service capacity, in detail.
Chapter 4: DSCR and LLCR: Evaluation of Debt Service Capacity
4.1 Definition of Debt Service Coverage Ratio (DSCR) and Its Role as Financial Covenant
The debt service coverage ratio is the financial indicator that banks prioritise most in project finance. DSCR is the abbreviation for Debt Service Coverage Ratio, indicating how many times the cash flow available for debt service in each period covers the debt service amount. The basic definition formula is as follows:
DSCR = CAFDS / DS
Where CAFDS is Cash Available for Debt Service, meaning cash flow available for debt service. DS is Debt Service, meaning principal and interest repayment for the current period. When DSCR falls below 1.0, cash flow is insufficient for debt service, which materialises as default risk.
The essential role of DSCR is to function as a financial covenant. Banks set minimum DSCR levels in loan agreements, and when those levels are breached, events of default are triggered. Table 4-1 shows DSCR’s function as a financial covenant.
Table 4-1: DSCR Financial Covenants and Bank Response Measures
| DSCR Level | Covenant Type | Legal Consequence | Bank Action |
| ≥ 1.30 | Target DSCR | Normal operations | No restrictions |
| 1.20 – 1.30 | Monitoring threshold | Partial dividend restriction | Enhanced monitoring |
| 1.15 – 1.20 | Minimum DSCR | Dividend cap at 50% | Cash sweep activation |
| 1.10 – 1.15 | Lock-up DSCR | Dividend prohibition | Reserve accumulation |
| < 1.10 | Event of Default | Loan acceleration rights | Step-in rights, management replacement |
As this table shows, DSCR is not merely an evaluation indicator but also a contractual threshold with legal consequences, including dividend restrictions, Cash Sweep activation, and default recognition. DSCR 1.20 is often set as the minimum standard to secure a 20% downside buffer. When DSCR falls below 1.15, the Lock-up clause is triggered, and dividends are entirely prohibited. Below DSCR 1.10, an Event of Default is triggered, and banks gain the right to demand accelerated loan repayment.
4.2 Industry-Specific DSCR Standards and Their Economic Rationale
DSCR standards vary significantly across industries based on risk profiles. Table 4-2 presents standard DSCRs for major industries and their rationale.
Table 4-2: Industry-Specific Standard DSCR Standards and Setting Rationale
| Industry | Minimum DSCR | Rationale | Revenue Stability |
| Toll Roads (Government Guaranteed) | 1.15-1.20 | Government support, stable traffic | Very high |
| Renewable Energy (FIT) | 1.15-1.20 | 20-year fixed tariff | Very high |
| Thermal Power (PPA) | 1.20-1.25 | Long-term PPA, fuel price risk | High |
| LNG Terminals | 1.25-1.30 | Partial spot market exposure | Medium-high |
| Oil & Gas E&P | 1.30-1.50 | Commodity price volatility | Medium |
| Mining | 1.40-1.60 | High geological and political risk | Low-medium |
Toll roads and renewable generation have minimum DSCR requirements of 1.15-1.20, relatively low, because long-term contracts or government guarantees stabilise revenue. Under FIT systems for solar power, a 20-year fixed-price purchase is guaranteed, so revenue predictability is extremely high, and 1.15 provides sufficient safety margin. Thermal power has fuel price volatility risk, but power purchase agreements guarantee revenue, so the level is 1.20-1.25.
LNG receiving terminals have long-term sales contracts but partially depend on spot markets, so a somewhat higher level of 1.25-1.30 is required. Oil and gas development and mining face diverse risks, including resource price volatility, reserve uncertainty, and political risk, so high levels of 1.30-1.60 are set. This high level is to ensure that debt service remains possible even if resource prices fall 30-50%.
4.3 Cash Sweep and Substantial Shortening of Repayment Period
In practice, flexible prepayment through Cash Sweep clauses is emphasised more than fixed repayment schedules. Cash Sweep is a mechanism that mandatorily applies excess cash flows to debt prepayment. Table 4-3 shows the major types of Cash Sweep.
Table 4-3: Cash Sweep Types and Trigger Conditions
| Cash Sweep Type | Trigger Condition | Sweep Amount | Economic Effect |
| DSCR Sweep | DSCR falls below target (e.g., 1.25) | CAFDS exceeding target DS × 1.25 | Maintains minimum DSCR |
| Sculpting Sweep | Throughout project life | Adjusted to maintain constant DSCR | Maximises minimum DSCR |
| Full Sweep | DSCR below lock-up level | 100% of available cash | Rapid deleveraging |
| Partial Sweep | DSCR in mid-range | 50-80% of excess cash | Balanced approach |
DSCR Sweep is the most common form; when DSCR falls below a certain level, all cash flows exceeding the amount needed to maintain the target DSCR are applied to prepayment. For example, if the target DSCR is 1.25, current-period CAFDS is 1,000 million USD, and DS is 672 million USD, the 160 million USD exceeding DS×1.25=840 million USD is applied to prepayment. Through this mechanism, DSCR is automatically maintained at 1.25.
Sculpting Sweep is a method that adjusts repayment amounts to maintain a constant target DSCR throughout the project lifecycle. By accelerating repayment in periods with abundant cash flows and easing repayment in periods with declining cash flows, DSCR is levelled. Through this method, the Minimum DSCR is maximised, and default risk is minimised.
The economic effect of Cash Sweep is a substantial shortening of the effective repayment period. Even with an initial 15-year repayment period, Cash Sweep often results in full repayment in 8-10 years. This shortening reduces total interest paid and improves shareholder returns. Therefore, scrutinising Cash Sweep conditions and their effects is more practically important than detailed repayment schedules.
4.4 Loan Life Coverage Ratio (LLCR) and Evaluation by Payback Period
The loan life coverage ratio is an indicator evaluating debt service capacity throughout the project lifecycle. LLCR stands for Loan Life Coverage Ratio and indicates how many times the present value of cash flows over the entire loan period covers the remaining debt. The definition formula is as follows:
LLCR = (PV of CAFDS from Evaluation Date Onward + Reserve Balances) / Remaining Debt at Evaluation Date
The importance of LLCR is that while DSCR is a flow indicator for each period, LLCR is a stock indicator for the entire future. Even if DSCR is temporarily high, if future cash flows are expected to decline, LLCR declines. Table 4-4 shows the essential differences between LLCR and DSCR.
Table 4-4: Essential Differences Between LLCR and DSCR
| Aspect | DSCR | LLCR |
| Time Horizon | Single period (annual) | Entire remaining loan life |
| Indicator Type | Flow indicator | Stock indicator |
| Focus | Current debt service capacity | Future aggregate capacity |
| Sensitivity | Immediate cash flow changes | Long-term trend changes |
| Bank Use | Operational monitoring | Strategic risk assessment |
| Typical Threshold | Minimum 1.20x | Minimum 1.30x |
Banks prioritise LLCR as a forward-looking indicator and utilise it for proactive risk management. When LLCR shows a trend of falling below 1.30, banks require business plan revisions or risk mitigation measures implementation.
In practice, Payback Period becomes an important evaluation indicator alongside LLCR. The Payback Period is the time required to recover the initial investment. In project finance, the debt Payback Period and the equity Payback Period are evaluated separately. Table 4-5 shows the Payback Period evaluation criteria.
Table 4-5: Payback Period Evaluation Criteria
| Payback Type | Measurement | Standard Criterion | Interpretation |
| Debt Payback | Years to accumulate CAFDS = Initial Debt | 60-70% of loan tenor | Back-end buffer for lenders |
| Equity Payback | Years to accumulate dividends = Initial Equity | 40-60% of project life | Front-loaded returns for sponsors |
| Example: 15-year Loan | – | 9-11 years | 4-6 years buffer period |
| Example: 25-year Project | – | 10-15 years | 10-15 years pure profit period |
The debt Payback Period being within 60-70% of the loan tenor is the bank’s standard requirement. For a 15-year loan, cash flows must accumulate to the debt principal within 9-11 years. Through this criterion, the latter half of the loan period becomes a margin period, securing a buffer against forecast errors and downside risks.
The equity Payback Period is the period required to recover initial equity through cumulative dividends. For a 25-year business, recovery within 10-15 years is the standard criterion. By meeting this criterion, shareholders can enjoy dividends in the remaining period as pure returns. Payback Period evaluation complements detailed time-series analysis of DSCR and LLCR, serving as a useful means to intuitively grasp project economics. The next chapter details the complex mechanism of dividend determination premised on these debt service capacity indicators.
Leave a Reply