The cruciality of dynamic, risk-based pricing in infrastructure

Credit ratings that convey the PD evaluate the ability and willingness of a borrower to repay debt obligations in full and on time. Such ratings range from ‘AAA’, indicating lowest PD, to ‘D’, indicating a default already, or near default.

LGD represents the loss that investors or lenders would incur should a debt instrument default. So, if a lender has an exposure of Rs 100 crore, an LGD of 60% would indicate the lender could suffer a Rs 60 crore loss in the event of default.

Both PD and LGD are dynamic and can vary based on factors such as sectoral, asset or instrument-specific, regulatory and macro.

To wit, debt instruments issued by two infrastructure projects in a similar type of business can have different PD ratings depending on the issuer’s leverage profile. The instruments can also have different LGDs depending on the proportion of capacity tied up with long-term contracts, or the presence of a force majeure clause in the concession agreement.

Hence, the assessment of both PD and LGD should be fundamental and dynamic, as a study by Crisil Ratings shows.

Evidence of improvement in infrastructure assets

Better risk sharing between public and private counterparties, greater presence of central counterparties to address payment issues, the Insolvency and Bankruptcy Code and earlier platforms for resolution of non-performing assets, and newer platforms such as infrastructure investment trusts (InvITs) for improving the availability and recycling of capital have led to a perceptible shift in the infrastructure sector’s credit profile.

For instance, the median PD rating for infrastructure assets in the Crisil Ratings portfolio has improved three notches from ‘CRISIL BBB+’ in 2018 to ‘CRISIL A+’ in 2024, implying lower credit risk in terms of likelihood of default.

LGD for infrastructure projects also show improvement.

Crisil Ratings studied nearly 150 stressed infrastructure assets of the past decade to understand the LGD trends in the sector, which pegged their LGD at 20-60% — well below the typical 60-65% factored in by lenders. The LGDs also varied across sub-sectors.

Infrastructure LGDs also show a downward trend over time, particularly after the implementation of the Insolvency and Bankruptcy Code, 2016. For instance, road sector LGDs declined to 20-25% against 35-40% before the code.

The conclusions are based on available data from a sample of reliable cases and may suffer from sampling bias. Also, some of the segments have very small samples to generalise conclusions. Further, the study spans the past decade and comprises a diverse range of projects, including projects that defaulted a long time back (legacy defaults) and those that defaulted recently. So, the results underscore transitional LGDs. The study’s findings must be interpreted in the context of these limitations.

Thus, LGD, like PD, is not constant and should be looked at more fundamentally.

Takeaway for stakeholders

Infrastructure assets demonstrate favourable LGDs compared with general corporate loans owing to the presence of structural safeguards with equitable risk sharing in contracts, characteristics such as simple nature of operations and strong investor demand after the emergence of InvITs, which has supported asset monetisation.

Given the large capital requirement of the infrastructure sector, it is critical to look at LGDs granularly. This can play a differentiating role in price determination, along with PD-based credit ratings.

While the PD ratings are reviewed and tracked periodically thanks to the long and established record of credit ratings, data availability on LGD remains a challenge. This has been the case globally as data on post-default recoveries are typically not disclosed publicly by stakeholders.

Given this, lenders tend to factor in high LGD estimates, with 60-65% the typical assumption in India. This estimate may not hold true always and needs to be viewed from a nuanced lens.

Factoring a reasonable LGD estimate is critical because a higher LGD can increase the cost of funds, which is not desirable for a sector instrumental for economic growth and with massive funding requirements. Similarly, a lower LGD estimate may underestimate risks.

In this context, embracing dynamic risk-based pricing, which captures the sectoral nuances of the various asset classes, can help mobilise funds at an optimal cost.

Lenders and investors may also consider using expected loss (EL) ratings, along with PD ratings.

Arithmetically expressed, EL = PD x LGD estimates. It is better placed to capture credit risk holistically.

Designed specifically for infrastructure projects at the behest of the Ministry of Finance, EL ratings can act as an additional input for lenders, investors, and issuers for pricing credit based on both pillars of credit risk — PD and LGD.

(Subodh Rai is Managing Director, Crisil Ratings Limited.)

Views are personal, and do not represent the stance of this publication.