On Aug. 9, 2022, S&P Global Ratings published a request for comment (RFC) on its proposed criteria, "Request For Comment: Project Finance Rating Methodology."

Following feedback from market participants, we finalized and published our criteria on Dec. 14, 2022, under two separate articles, "General Project Finance Rating Methodology," and "Sector-Specific Project Finance Rating Methodology." We also finalized and published the sector and industry variables report, "Sector And Industry Variables: Project Finance Rating Methodology." A sector and industry variables report (SIVR) is a publicly available criteria-related publication that describes sector, industry, asset class, or regional variables that we expect to periodically update--mainly to reflect our views on changing macroeconomic and market conditions. The SIVR is not criteria, but it is intended to be read in conjunction with the criteria.

We'd like to thank investors, issuers, and other intermediaries who provided feedback. This RFC Process Summary provides an overview of the external written comments and certain other feedback we received on the proposed criteria, the analytical changes we made following the RFC period, and the rationale for those changes. There's also a summary of the changes we made to the SIVR following the RFC period.

External Written Comments Received From Market Participants That Led To Significant Analytical Changes To The Final Criteria

We did not receive any external written comments from market participants that led to significant analytical changes to the criteria.

External Written Comments Received from Market Participants That Did Not Lead To Significant Analytical Changes To The Final Criteria

This section addresses written comments or questions that did not lead to a change or that led to edits that do not involve significant analytical changes to the final criteria.

One respondent requested clarification on the treatment of letters of credit (LOCs) in our analysis. We amended the relevant section in the criteria to specify that the treatment depends on who is responsible for the repayment, and we include in our debt service calculations the costs related to an LOC only if it is an obligation of the project. We do not view this addition as a significant analytical change.

One respondent inquired about our rationale for using the median debt service coverage ratio (DSCR) as opposed to the mean when performing the resiliency assessment. We aim to improve consistency and comparability across projects and we believe using the median provides a better comparison in our analysis, because it is not distorted by potential outliers in one or more periods in our forecast. We also expect to use our analytical judgement when assessing a project's resiliency: for example, if the median DSCR is high but the trend in the DSCR is declining, we may not apply the adjustment. We did not modify the criteria as a result of this question.

One respondent requested clarification on our treatment of projects that are acquiring newly built assets, rather than building them. Our rating framework is clear that if the acquired assets are already operating, we consider those as part of the operations phase. If the assets acquired bear material construction risk, we assess their performance and risks in the construction phase, as well as the operations phase, and determine the project stand-alone credit profile (SACP) by considering the lower of the operations phase and construction phase SACPs. However, if the project acquires fewer assets than originally expected, our analysis only considers the acquisitions that actually materialize. We haven't modified the criteria to reflect this comment because we believe the criteria framework appropriately covers these situations.

One respondent inquired our rationale for using a five-year horizon for the market exposure case assessment, as opposed to a longer period or even the life of the rated debt. We believe a time horizon of up to five years is an appropriate time frame to model credit, market, and operational risk exposures, based on our observations of the history of market fluctuations and cycles. We did not change the criteria because of this question.

One respondent inquired about the effect of moving toward a more data-driven approach for the liquidity assessment during the construction phase. We do not expect this move to affect construction phase SACP outcomes, and we believe our revised approach is more transparent and fosters more consistency in our analysis of liquidity during construction. Hence, we did not change the criteria following this question.

One market participant commented that the criteria did not explain in detail how we consider external guarantees and government support. We did not change the criteria to reflect this comment because these analytical factors are detailed in our related methodologies, which are listed in the criteria.

When we determine the counterparty dependency assessment (CDA) for a construction counterparty, our criteria consider the liquidity available for replacement. One respondent commented that they would also include nonliquid credit enhancements. As explained in our framework, our replaceability analysis focuses on liquidity levels, which need to be sufficient to ensure cash flows are not disrupted and construction continues. We do not view nonliquid credit enhancements as providing similar protection, so we did not change the criteria as a result of this comment.

We received comments related to the introduction of the future value modifier. One respondent welcomed the approach and suggested reducing the requirements to apply the modifier to projects with low volatility, while another respondent questioned the introduction of the modifier in our methodology. Our objective in adding the future value modifier is to recognize the level of additional flexibility provided by overcollateralization in a transaction that has a material tail after the debt matures, which the minimum DSCR analysis does not capture. We believe that the more value a project can generate above the value of its debt, the more resilient it is, and the more flexibility it has to withstand unforeseen stresses. In our view, this improves not only the project's ability to refinance and its recovery prospects (which are captured elsewhere in our analysis), but also the project's ability to renew contracts, negotiate with lenders, raise additional debt or equity if needed, etc. We require a minimum tail of least 10 years to make a positive adjustment, because we want to recognize projects that have superior credit strength compared with peers. We believe the introduction of this modifier refines our analysis and we did not change the criteria following these comments.

In relation to the debt structure adjustment section, one respondent requested clarification of our definition and treatment of unusually high mandatory payments in later years. We use our analytical judgement and do not refer to a specific threshold to make that determination. Instead, we analyze comparable transactions to determine if the amortization payments in the last years of a project are unusually high, particularly if coverage ratios rely more heavily on growth assumptions. We did not change the criteria following this comment.

Other Substantive Market Feedback That Did Not Lead To Significant Analytical Changes To The Final Criteria

This section summarizes certain other feedback from market participants that did not involve significant analytical changes to the final criteria.

One respondent requested clarification of our requirements for assigning a strong liquidity assessment when determining the operations phase SACP. We received another comment regarding the potential to double-count the benefit provided by reserves in our liquidity calculations. As a result of these comments, we updated the liquidity section as follows.

• We clarified our definition of sources of liquidity: we specified that, on top of cash flow available for debt service (CFADS), sources of liquidity include reserves when designated as available for debt repayment in the transaction documents, proceeds from asset sales, and undrawn committed credit facilities. They may also include cash that is trapped in the structure if it cannot be distributed and is available to repay debt. We also clearly specified that sources would not include those reserves to which we already give credit in the analysis--for example, in our resiliency assessment--or any reserve balance that is reserved for a specific expense in our analysis, such as a capital expenditure reserve. This ensures that we would not count the same liquidity source twice.
• We reviewed our conditions for assessing liquidity as strong and introduced, on top of the ratio of sources to uses of liquidity, an additional condition based on the size of reserves available to repay debt and the potential volatility of CFADS. We see this change as a fine-tuning of our analysis, rather than as a substantive change to our analytical approach.

Following one respondent's suggestion, we clarified in the criteria that where projects have debt that rely on economic or operational tests to trigger debt repayment, instead of having fixed contractual maturity dates, they do have a final legal maturity date. We do not view this clarification as a significant analytical change.

We received a comment about our recalibration of the preliminary operations phase SACP table (table 15) in the operations methodology. The new calibration is intended to improve rating transitions in situations when the operations phase business assessment (OPBA) changes from one category to another, for instance, when the OPBA moves to 5-6 from 3-4. The new thresholds incorporate data collected on our portfolio of rated transactions since the previous criteria was published eight years ago. We haven't made any changes to the criteria as a result of this comment.

One respondent commented on the limited transparency of our approach to forecasting macroeconomic assumptions. We explain how we forecast those variables in the section where we describe our base-case and downside-case scenarios, complemented with our SIVR. This applies to several macroeconomic variables, including foreign exchange, interest rates, and inflation. We do not think we need to expand the criteria on this aspect.

Significant Analytical Changes To The Final Criteria That Did Not Arise From Market Feedback

We finalized and published the final criteria without making any significant analytical changes to the criteria that were unrelated to the market feedback we received.

However, as a result of further testing after the publication of the RFC, we introduced minor recalibrations in two tables that we believe increase the consistency of application without substantially changing our approach.

• In table 5, we recalibrated the thresholds we use to assess market exposure, to allow for more judgement when the decline of CFADS from the base case to the market exposure case is close to the boundaries between each category. We also further explained the impact of a project's competitive position when determining the market risk score.
• In table 22, we revised upward some of the thresholds for the sources-over-uses ratio used to determine the CDA for a construction counterparty, to better align with the characteristics of our portfolio.

We also incorporated a few editorial revisions, where appropriate, to improve the readability and clarify the intent of the criteria. For example, in the construction phase business assessment (CPBA) section, we edited the text in various places to clarify the impact of the various modifiers on the CPBA score. Similarly, in both the construction and operations holistic analysis sections, we specified that a holistic adjustment cannot override any cap that applies to our ratings. In addition, we harmonized some wording and aligned certain references in tables and charts (including tables 2, 7, 9, 16, 21, 23, and 49, and chart 27).

We also provided the following clarifications and amendments based on internal deliberations, none of which significantly alters the analytical approach on an individual or cumulative basis:

• In table 9, when assessing resiliency, we amended the criteria to reword a sentence that misrepresented our intent. We now state that stronger liquidity reserves should represent at least one year of debt service or 5% of total outstanding project debt.
• We explained further how we treat additional debt in our analysis of limited-purpose entity covenants. In particular, we specified in the structural protection section that we generally assume all permitted debt is drawn in our base-case and downside forecasts, unless we have evidence, a track record, or another strong reason, to expect that it will not be drawn.
• We clarified the application of the methodology for subordinated debt in three places. First, in the section "Calculating The Minimum Debt Service Coverage Ratio," when considering whether to exclude a DSCR period from the minimum DSCR calculation for applying table 8, we clarified that we may consider excluding a period separately for the senior and the subordinated tranches. Second, in the liquidity section, we specified that liquidity sources for subordinated debt are net of senior debt service, reserve replenishment, and any other uses that are senior to subordinated debt service in the waterfall. Finally, in the glossary, we specified that the project life coverage ratio for a subordinated tranche of debt is calculated using the CFADS available to that tranche of debt and outstanding debt at that seniority of debt.
• In the counterparty section, we made it clear that our analysis of liquidity for replacement would include only funds that are available to offset the specific risk that the counterparty may fail, and that are not earmarked for another purpose.
• We amended the construction phase business assessment section to make it clear that project-specific attributes, such as technology and design, can worsen but not improve the construction difficulty score. This is corroborated by the paragraphs in that same section that detail how these factors can affect the difficulty of construction.
• In applying tables 17, 18, and 22, we stipulated that the ranges include the value at the lower boundary, but not the upper boundary.
• We clarified that, for long construction periods, or if we believe that the performance of a construction counterparty can affect the project's creditworthiness before replacement, we might update the CDA during the surveillance process.
• We detailed further the conditions under which a project that is exposed to a possible tax consolidation of its parent may not be capped by the creditworthiness of its parent.
• We expanded the section related to debt issued by project finance holding companies, to provide details on our specific approach for holding companies that own a single encumbered operating company, versus those that own multiple operating entities.
• The only change we made related to our "Sector-Specific Project Finance Rating Methodology" was to clarify a few assumptions in the tables that had been numbered 35, 38, 43, 46, and 47 in the RFC. We also clarified that the assumptions used are typical for our analysis of these sectors.

Significant Changes To The Sector And Industry Variables Report

We finalized and published the SIVR without making any significant changes to the version published in the RFC.

We published the SIVR separately from the two criteria articles, "Project Finance Rating Methodology" and "Sector-Specific Project Finance Rating Methodology." The report is not criteria, but it is intended to be read in conjunction with the criteria. For further information, see "Evolution Of The Methodologies Framework: Introducing Sector And Industry Variables Reports," published Oct. 1, 2021