Analysis of Debt Spreads
This research analyzes the pricing of credit risk in mortgage loan markets over the period 1996 through Q3 2025.
It focuses on how credit spreads have evolved across different levels of borrower leverage and how those spreads
compare to risk-free benchmarks. The objective is to understand how investors have priced credit risk in the
mortgage sector under varying market conditions, including both stable periods and times of financial stress. By
examining these patterns over nearly three decades, the analysis provides insight into the structure of credit
markets, shifts in risk tolerance, and the impact of broader economic cycles on the pricing of mortgage credit.
Credit Spread: 75% LTV – 25% LTV
Illustrates how the credit premium between high-leverage (75%) and low-leverage (25%) mortgage loans has changed over time. Use the dropdown menu to toggle between results computed over the full sample period and those excluding the Global Financial Crisis (GFC) years.
Credit Spread: 75% LTV – Risk-Free Rate
Illustrates how the credit premium between high-leverage (75%) mortgage loans and Risk Free Rate has changed over time. Use the dropdown menu to toggle between results computed over the full sample period and those excluding the Global Financial Crisis (GFC) years.
METHODOLOGY:
-
This analysis investigates the relationship between loan-to-value (LTV) ratios and gross credit spreads in the commercial mortgage market, using quarterly data from 1996 through Q3 2025. For each quarter, observed gross spreads are paired with their corresponding LTV levels. To estimate the pricing curve of credit risk, a linear regression is applied using a transformed version of the LTV.
-
The independent variable is calculated as the leverage ratio, defined as LTV / (1 − LTV), to capture the nonlinear increase in credit risk as borrower leverage rises. The dependent variable is the gross credit spread, measured in basis points. A base point is included at LTV = 0% with an assumed spread of 30 basis points, representing structural features unrelated to credit risk such as servicing costs, origination fees, and liquidity premiums.
-
The regression model is specified as:
Spread = β₁ × (LTV / (1 − LTV)) + β₀
Where:
• β₁ is the slope coefficient, capturing the incremental spread attributable to rising leverage.
• β₀ reflects the baseline spread from structural costs.
-
The fitted equation is used to estimate gross spreads at standard LTV thresholds. These estimates are then added to the contemporaneous risk-free rate (e.g., Treasury or swap rate) to compute the implied mortgage interest rate. This approach enables consistent time-series analysis of how markets have priced credit risk at varying leverage levels.
To illustrate the time-series behavior of spreads, two measures were computed each quarter: the spread between estimated gross spreads at 75% and 25% LTV, and the spread between the 75% LTV estimate and the corresponding risk-free rate. These spread differences were plotted over time to observe fluctuations in the credit curve. To summarize their behavior, average and standard deviation bands were also plotted. In these charts, users can choose to exclude the Global Financial Crisis period (2008–2010) using the dropdown at the top of the graph. Excluding this period removes crisis-driven volatility from the average and standard deviation calculations, allowing the underlying structural trends to be seen more clearly..
DATA SOURCES:
-
Commercial mortgage commitments data from the American Council of Life Insurers (ACLI) covering Apartments, Office, Retail, and Industrial properties (1996 to Q3 2025).
-
Market yield data from the Federal Reserve Economic Data (FRED) database, maintained by the Federal Reserve Bank of St. Louis.
These analyses are intended solely for academic purposes. No warranty or representation is made with regard to their accuracy.