Securitization and Credit Quality

1. Introduction & Research Question

The paper Securitization and Credit Quality investigates how securitization affects the subsequent credit performance of corporate loans. In particular, the authors ask whether Securitization and Credit Quality are negatively linked—i.e., whether securitized loans perform worse (in terms of default risk) than comparable non-securitized loans, and under what conditions.

The motivation is grounded in informational asymmetries: banks typically have better information about the borrowers they originate and more incentives to monitor them. When a bank securitizes a loan (i.e. sells it into the capital markets via collateralized loan obligations, CLOs), the bank’s incentive to monitor could be weakened. Thus, Securitization and Credit Quality are possibly connected via a monitoring channel.

The authors focus on the euro-denominated syndicated corporate loan market. They assemble a detailed dataset identifying which loans are securitized (via CLOs) and which are not, and then compare credit quality evolution over time between securitized and non-securitized loans. Their central finding is that while at issuance there is little evidence that banks systematically securitize lower credit quality loans, over time securitized loans tend to deteriorate more — thus indicating a negative effect on credit performance. That is the core of their evidence on Securitization and Credit Quality.

Securitization and Credit Quality

They also examine how this effect varies by collateralized vs non-collateralized loans, and whether the deterioration is consistent with a reduced monitoring incentive hypothesis. The paper contributes to the ongoing debate about whether securitization leads to moral hazard, weaker screening, or adverse selection.

Banks are usually better informed on the loans they originate than other financial intermediaries. As a result, securitized loans might be of lower credit quality than otherwise similar nonsecuritized loans. We assess the effect of securitization activity on loans’ relative credit quality by employing a uniquely detailed dataset from the euro-denominated syndicated loan market. We find that, at issuance, banks do not seem to select and securitize loans of lower credit quality. Following securitization, however, the credit quality of borrowers whose loans are securitized deteriorates by more than those in the control group. We find tentative evidence suggesting that poorer performance by securitized loans might be linked to banks’ reduced monitoring incentives. Banks generate proprietary information and tend to have superior knowledge on the credit quality of the loans they originate. As a result, banks might have an incentive to securitize loans of lower credit quality to unsuspecting investors (Gorton and Pennacchi, 1995). Largely for this reason, securitization has been perceived as a major contributing factor to the 2007-2009 financial crisis (Financial Crisis Inquiry Commission, 2011). Following the crisis, authorities have investigated a number of banks over claims related to securitized loans.1 In this direction, recent empirical evidence suggests that banks tend to securitize the riskier mortgages of their portfolio (see for instance Krainer and Laderman, 2014; Elul, 2015). Yet, evidence on the impact of securitization on corporate loans on credit quality remains very limited, it is circumscribed to the U.S. and seems to offer contradictory results (Benmelech et al., 2012; Bord and Santos, 2015).

2. Literature & Hypotheses

2.1 Existing Evidence & Theoretical Mechanisms

The authors situate their work in two strands of literature:

  1. Studies on banking, risk taking, and credit cycles, which suggest that securitization may weaken incentives for screening and monitoring, thus potentially harming credit quality. (E.g. Gorton & Pennacchi, Petersen & Rajan)

  2. Empirical work on securitization and performance of securitized loans. Some U.S. studies find underperformance of securitized loans (Benmelech et al.), while others do not.

Within that debate, Securitization and Credit Quality could be influenced by two (not mutually exclusive) forces:

Thus, the authors propose three hypotheses:

Testing these is central to uncovering the nature of Securitization and Credit Quality.

3. Data & Methodology

3.1 Dataset & Sample Construction

3.2 Empirical Strategy

The core empirical design is a difference-in-differences / interaction regression:

ΔCreditRiski,t=α+β⋅Securitizedi  +  γ⋅Controls  +  FixedEffects  +  εi,t\Delta \text{CreditRisk}_{i,t} = \alpha + \beta \cdot \text{Securitized}_i \;+\; \gamma \cdot \text{Controls} \;+\; \text{FixedEffects} \;+\; \varepsilon_{i,t}

More precisely, they estimate specifications where the change in EDF (expected default frequency) over some horizon (e.g. 1, 2, 3 years) is regressed on whether the loan was securitized, interacting with time, and controlling for characteristics. IMF+2ResearchGate+2

They also use logistic regressions to test whether banks select lower or higher credit risk loans for securitization (i.e., comparing observable credit quality at issuance). IMF+2SSRN+2

To alleviate endogeneity, they implement propensity score matching (PSM): for each securitized loan they find a matched non-securitized loan with similar observable characteristics, and then compare performance. This approach helps isolate the effect of securitization on credit quality trends. IMF+2SSRN+2

They further split the sample by collateralized vs non-collateralized loans (to test if the monitoring effect is more pronounced when collateral is absent). IMF+2SSRN+2

Lag structures, alternative control sets, robustness checks (placebos, alternative variable definitions), and subsample sensitivity are also used to validate findings.

4. Results & Evidence on Securitization and Credit Quality

4.1 At Issuance: Selection Patterns

On the question of selection (H1), the authors find:

Thus, initial evidence does not support that securitization is used to offload bad loans exclusively. Instead, selection seems favorable.

4.2 Post-Securitization: Deterioration in Credit Quality

Turning to H2 and H3 — how credit risk evolves after securitization — the results are more striking:

These findings provide strong support for the notion that Securitization and Credit Quality are negatively related via post-origination deterioration.

4.3 Collateralized vs Non-Collateralized Loans

To test whether the monitoring channel is stronger when loans are not backed by collateral, the authors split the sample:

Hence, Securitization and Credit Quality seem more tightly linked (negatively) in non-collateralized segments.

4.4 Robustness & Sensitivity

To bolster the results, the authors conduct multiple robustness checks:

These robustness checks strengthen the credibility of the finding that Securitization and Credit Quality are negatively associated, especially after securitization.

5. Interpretation & Mechanisms

Given the empirical results, the authors discuss plausible explanations for the negative link between Securitization and Credit Quality post-issuance.

5.1 Weakening Monitoring Incentives

The leading interpretation is that after securitization, the originating bank has less incentive to monitor the borrower, because the loan is off its balance sheet. This monitoring lapse leads to looser oversight, higher risk-taking or deteriorating performance. The stronger effect for non-collateralized loans supports this: when collateral is absent, monitoring is more critical, and its weakening leads to worse outcomes.

Thus, Securitization and Credit Quality are connected via a moral hazard channel: banks may tend to “pass the buck” after securitization, letting performance suffer.

5.2 informational advantage and the “Lemons” Hypothesis (less supported)

Another possibility is the “lemons” hypothesis: that banks exploit their informational advantage by retaining good loans and securitizing riskier ones (i.e. hiding lemons). But the evidence at origination supports the opposite: securitized loans are of better, not worse, credit quality. Hence, the authors find less support for the lemons story. The observed deleterious effect is more consistent with Securitization and Credit Quality being harmed through weakening monitoring rather than adverse origination selection.

5.3 Dynamic / Incentive Effects & Reputation

Banks may balance short-term gain (selling loans) and reputation costs (if losses are high). Over time, delayed costs may emerge, but if monitoring is weak, deterioration can still happen. The paper suggests that reputation and long-term relationships may moderate but not fully offset the deterioration. Also, during crisis periods or in downturns, the negative feedback is magnified.

6. Policy & Stability Implications

This study’s findings on Securitization and Credit Quality carry important implications for financial regulation, risk management, and market design.

6.1 Regulatory Safeguards & Monitoring Incentives

If securitization reduces monitoring and leads to worse credit quality, regulation should aim to realign incentives:

6.2 Collateral & Lending Standards

Given that the negative effect is stronger for non-collateralized loans, regulators might require stricter screening or oversight for unsecured securitization. Collateralized securitization appears safer under this mechanism.

6.3 Crisis Amplification & Procyclicality

Because Securitization and Credit Quality diverge more negatively in crisis periods, securitization may amplify downturns. Loans securitized before downturns may underperform more, feeding into banking stress, defaults, and systemic risk. Thus, securitization markets may contribute to procyclical behavior and exacerbate financial cycles.

6.4 Designing Better Securitization Markets

To mitigate negative effects:

Thus, understanding Securitization and Credit Quality is vital for resilient securitization markets.

7. Limitations & Directions for Future Research

The authors acknowledge limitations and suggest paths forward:

Future research could expand to other markets, more recent periods, compare mortgage vs corporate securitization, or integrate contract-level features (e.g. covenants, servicer incentives) to better trace the channels linking securitization to credit outcomes.

8. Summary & Key Takeaways

Below is a concise recapitulation of the main conclusions, with focus on the term Securitization and Credit Quality:

  1. The paper Securitization and Credit Quality explores how securitization affects subsequent credit performance of corporate loans, focusing on European CLO markets.

  2. At loan issuance, banks appear to securitize better, not worse, credit quality loans (supporting a signaling selection hypothesis).

  3. Over time, securitized loans experience worse deterioration in credit quality (higher default risk) relative to non-securitized loans, especially during crisis periods.

  4. The negative effect is more pronounced among non-collateralized securitized loans, consistent with a monitoring incentive channel.

  5. Robustness tests (propensity score matching, alternative specifications) affirm the core result: Securitization and Credit Quality are negatively related post-securitization.

  6. The authors argue that weakened monitoring after loan transfer is the main mechanism, rather than adverse initial selection.

  7. Policy implications include retention requirements, enhanced disclosure, stricter oversight for unsecured securitization, and design improvements to reduce procyclical risk.

  8. Limitations include sample scope, unobserved confounders, and coarse collateral data; future work could extend to broader samples and more granular data.

Conclusion

In short, Securitization and Credit Quality are tightly linked: while securitization may not degrade credit quality initially, it tends to allow deterioration over time, especially when monitoring is weak. This result has serious implications for securitization design, regulation, and systemic risk management.

Also Read: The Residual Income Approach to Housing Affordability: The Theory and the Practice