## Macro-Financial Linkages and Heterogeneous Non-Performing Loans Projections: An Application to Ecuador

_IMF Working Papers, December 7, 2016_

## Source details

**Canonical URL:** [Macro-Financial Linkages and Heterogeneous Non-Performing Loans Projections: An Application to Ecuador](https://www.imf.org/en/publications/wp/issues/2016/12/31/macro-financial-linkages-and-heterogeneous-non-performing-loans-projections-an-application-44446)

## Other formats

- [Markdown version](/en/publications/wp/issues/2016/12/31/macro-financial-linkages-and-heterogeneous-non-performing-loans-projections-an-application-44446/index.md)
- [Structured JSON version](/en/publications/wp/issues/2016/12/31/macro-financial-linkages-and-heterogeneous-non-performing-loans-projections-an-application-44446/index.json)
- [Bundle manifest](/en/publications/wp/issues/2016/12/31/macro-financial-linkages-and-heterogeneous-non-performing-loans-projections-an-application-44446/bundle-manifest.json)

## Bibliographic details
- Authors: Francesco Grigoli, Mario Mansilla, Martín Saldías
- Published: December 7, 2016
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9781475559347.001

---

### Summary of the framework
- Proposes a stress testing framework of credit risk that:
  - Analyzes macro-financial linkages.
  - Generates consistent forecasts of macro-financial variables.
  - Projects non-performing loans (NPL) on the basis of such forecasts.
- Key premise: Economic contractions are generally associated with increases in NPL.
- Challenges common empirical assumption of homogeneous impact across banks and instead:
  - Accounts for banks’ heterogeneous reaction to macro-financial shocks in a dynamic context.
  - Accounts for potential cross-sectional dependence across banks caused by common shocks.

### Major findings and application to Ecuador
- An application to Ecuador suggests:
  - Substantial heterogeneity is present across banks in how macro-financial shocks affect NPL.
  - Imposing homogeneity may lead to overestimating or underestimating the resilience of the financial system to macroeconomic shocks.
  - Heterogeneity should be taken into account when trying to anticipate inflections in the quality of loan portfolios.

### Methodology and technical features
- Analytical components:
  - Dynamic modeling of banks’ heterogeneous responses to macro-financial shocks.
  - Treatment of cross-sectional dependence across banks due to common shocks.
- Econometric and modeling tools referenced in the content and keywords:
  - Vector autoregression.
  - Panel data estimation.
  - Difference GMM estimator.
  - Likelihood-ratio test statistic.
  - Regression equation formulation for NPL projection.
- Subject areas listed: Banking, Credit, Econometric analysis, Financial crises, Financial institutions, Global financial crisis of 2008-2009, Money, Nonperforming loans, Oil prices, Prices, Vector autoregression.

### Policy relevance and stress-testing implications
- Stress-testing implications:
  - Stress tests that assume homogeneous bank responses to macro-financial shocks risk mismeasuring system resilience.
  - Incorporating bank-level heterogeneity and cross-sectional dependence leads to more accurate projections of NPL under macroeconomic stress.
- Practical recommendation:
  - When anticipating changes in portfolio quality, take into account bank-specific sensitivities to macro-financial variables rather than relying on uniform effects.

---

## Content in this bundle

- **1. Panel Regression Results**
  - [1. Panel Regression Results (Markdown version)](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16236.pdf.md){rel="alternate" type="text/markdown"}
  - [1. Panel Regression Results (PDF)](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16236.pdf){rel="external" type="application/pdf"}

---

_Source: https://www.imf.org/en/publications/wp/issues/2016/12/31/macro-financial-linkages-and-heterogeneous-non-performing-loans-projections-an-application-44446_