By Moody's Investors Service Associate Vice President, Wing Ng and Associate Analyst, Wei Hu
Changes over time in the weighted average coupon, or WAC, of a securitized pool of auto loans can have a significant impact on the amount of excess spread in the transaction. The amount of available excess spread is critical to assessing the total amount of credit enhancement protecting bondholders, both at the time a transaction is initially rated and over its life.
This article describes the common causes of WAC changes and how they may differ for pools of different credit qualities. It also reviews the findings of Moody's data analysis of WAC changes in over 250 Moody's-rated transactions that closed between January 1995 and December 2002, totaling over $250 billion. We find that:
1. The WACs of higher-credit-quality pools tend to increase over time, while the WACs of lower-credit-quality pools tend to decline;
2. For those pools that experienced increases in the WAC, the increases tended to be larger for higher-credit-quality pools;
3. For pools that experienced decreases in the WAC, the declines tended to be larger for lower-credit-quality pools.
These findings are consistent with the way Moody's incorporates pool credit quality in its assessment of the potential value of excess spread in providing credit protection over the life of an auto-loan-backed transaction.
How WAC changes can affect excess spread
Determining the value of excess spread as credit protection is an important aspect of analyzing any auto ABS transaction, because excess spread typically forms the investor's first line of defense against credit losses. Gross excess spread of an auto securitization is typically defined as the interest collected on the underlying auto collateral less the fees paid in the transaction and interest paid on the securities. The interest collected from the auto collateral is the aggregate of the interest component of the installment payments of the individual auto loans.
Changes in a pool's WAC refer to shifts in the average interest rate paid on the underlying pool of assets over the life of the securitization as various factors interact to increase or decrease the relative proportions of high- vs. low-APR contracts in the pool. WAC changes - both potential and actual - are an important consideration when estimating the level of excess spread that is available to cover credit losses in a transaction.
Common causes of WAC changes
Four main factors account for most WAC changes were observed in the auto ABS market. We describe each of them separately below:
* Prepayments - As with pools of mortgage loans, prepayments can affect the WAC of auto pools. In a decreasing interest rate environment, individuals with fixed-rate mortgage loans have a greater incentive to prepay (and refinance) their mortgages at the new (lower) market interest rate. Although auto portfolios do not exhibit the same strong correlation between prepayments and the current interest rate as mortgage portfolios, an auto portfolio can experience WAC deterioration like a mortgage pool if the underlying loans with higher APR are paid off at a faster pace than lower APR loans. On the other hand, prepayments can also cause the WAC to appreciate, as exhibited by several of the securitized prime auto pools. Lower-interest rate loans in auto pools are often associated with better credit obligors who are more capable of prepaying their loans. If that portion of the pool prepays faster than the remainder of the pool, the effect of those prepayments alone would cause the pool WAC to appreciate over time.
* Defaults - Higher interest rate loans in auto pools are often associated with weaker-credit obligors who are more likely to default on their loans. If that portion of the pool has a higher default rate than the remainder of the pool, the effect of those defaults alone would cause the pool WAC to depreciate over time.
* Chapter 13 bankruptcy filings - Chapter 13 filings may reduce both principal and interest payments of auto loans depending on a bankruptcy court order. Effectively, the bankruptcy court may restructure the loan and replace that loan with one that has a lower balance and/or lower APR. The lowering of the APR under a Chapter 13 filing has the effect of reducing the WAC on the portfolio when large numbers of such restructurings occur.
* APR distribution of loans - When the pool of underlying loans is weighted so that loans with higher APR have shorter terms than the pool's other loans, then the WAC will decrease over time as the pool amortizes. This happens because the high-coupon loans pay off more quickly and are retired early, while the lower coupon-paying loans remain outstanding longer. The converse can occur when the pool of underlying loans is weighted so that loans with lower APR have shorter terms than the pool's other loans. A securitized pool may have such distributions when an issuer combines new originations with more seasoned loans (e.g. "clean up call" loans from a matured securitization) that were originated when interest rate levels were substantially different.
Analysis of historical WAC changes
Moody's analyzed the historical WAC changes in Moody's-rated U.S. automobile loan-backed ABS deals that closed between January 1995 and December 2002, based on performance data through February 2005. Only public and 144A deals backed by amortizing pools (i.e., no revolving periods) were included in the analysis. Another selection criterion was to limit deals only to those that had amortized down to a pool factor of 15% or lower as of the last reporting date. Based on these conditions, the dataset used for the analysis consisted of 132 prime deals totaling $197 billion of issuance, 57 near-prime deals totaling $35 billion and 74 subprime deals totaling $32 billion.
We define the "WAC change" for each transaction as the difference between (1) the WAC reported prior to the exercise of a clean-up call on the transaction (estimated to equal the last WAC reported when the pool factor was greater than or equal to 15%) minus (2) the WAC reported on the first servicer report after deal closing. At the same time, the ultimate cumulative net loss performance of each pool was estimated by calculating the pool's cumulative net losses as a percentage of cumulative liquidations (or ending "loss-to-liquidation") for the same reporting date when the "ending WAC" was determined. The purpose of this last step was to classify deals into the prime, near-prime and subprime categories based on an objective measurement of credit quality.
The vast majority of near-prime and subprime pools have experienced WAC deterioration, while most prime pools have experienced WAC appreciation. The results of the WAC change analysis are consistent with the way Moody's incorporates pool credit quality in its assessment of the potential value of excess spread in providing credit protection over the life of an auto-loan-backed transaction.
Conclusion
An understanding of the potential direction and magnitude of WAC changes and the common causes of those changes are important in helping all parties to an auto ABS deal gauge the value of excess spread as credit protection against losses on the underlying loans. Changes to the WAC of pools underlying auto loan ABS transactions can have a material effect on the excess spread generated by the pools. Moody's research in this special comment summarizes the typical extent of those WAC changes. It also highlights the difference between subprime and near-prime pools, which typically exhibit WAC deterioration, and prime pools, which typically exhibit WAC appreciation.
One possible explanation for those differences is that defaults/bankruptcies occur more frequently with lower credit quality, higher APR borrowers while prepayment occurs more frequently with higher credit quality, lower APR borrowers. In other words, WAC deterioration in subprime pools may be driven by the high level of defaults/bankruptcies. Conversely, WAC appreciation in prime pools may be driven by the high level of prepayments
(c) 2005 Asset Securitization Report and SourceMedia, Inc. All Rights Reserved.
http://www.asreport.com http://www.sourcemedia.com
