Primary navigation:

QFINANCE Quick Links
QFINANCE Reference

Home > Financial Risk Management Best Practice > Factoring Problem Loan Assumptions into Asset–Liability Management (ALM) Modeling

Financial Risk Management Best Practice

Factoring Problem Loan Assumptions into Asset–Liability Management (ALM) Modeling

by Gary Deutsch

Executive Summary

  • Traditional ALM modeling considers interest rate risk assumptions for loans based on new loan funding amounts and subsequent payments, prepayments, and maturing balances.

  • However, it ignores modified payments, missed payments, and default issues such as impaired loans, repossessions, and foreclosures.

  • It is these problem loan issues that are the subject of this chapter. Incorporating problem loan data and assumptions into ALM modeling is an evolving topic. This chapter provides guidance.


An essential part of the interest rate risk (IRR) management process is the analysis and documentation of IRR assumptions. Since all methods used to analyze the rate sensitivity of financial instruments require estimates of the amount and timing of their cash flows, the embedded prepayment options found in many types of loans must also be estimated. These estimates must include the level and speed of prepayments for an expected rate environment. However, economic conditions may not always favor prepayments. Therefore, IRR assumptions should also include assumptions related to problem loans. For instance, late payments, restructured payments, and payments that will not be collected can have a significant impact on IRR assumptions, especially in more volatile economic times.

This chapter describes the types of assumptions related to problem loans that ALM managers should incorporate into the interest rate risk management process. Note that although the references in this chapter are to US Generally Accepted Accounting Principles (US GAAP), the IRR cash flow concepts should hold true for all financial institutions worldwide. For additional information on this topic, refer to the More Info section at the end of the chapter.

Industry Support

Including problem loan assumptions in IRR analyses is not a new idea, but it has not been widely adopted. Among the many post-Great Recession lessons learned is one that curiously few are discussing—the obvious need to better integrate credit risk assumptions with interest rate risk and liquidity risk in the asset and liability management process. According to Michael R. Guglielmo, managing director of the Darling Consulting Group in Newburyport, Massachusetts, the tools for ALM modeling often depend on the level of optionality embedded in an institution’s balance sheet. He adds that asset and liability management committee (ALCO) activities that were once thought to be “nice to have” have quickly become “have to have.” Among the long list of these activities is his belief that “credit and default-risk modeling will continue to migrate into ALM modeling.”1

The PricewaterhouseCoopers (PwC) 2009 survey of the balance sheet management practices at 43 leading financial institutions across the globe2 defines the four main areas of balance sheet management to be interest rate risk management, liquidity risk management, capital management, and management of discretionary investment portfolios. PwC concludes that the severe financial crisis in 2008 prompted banks globally to broaden their ALM perspective to include a more holistic view of their institution’s balance sheet, but credit risk management is not on the list. Although balance sheet management is a good progression in the natural extension of ALM, the prospect of continued worldwide economic volatility should encourage more ALM managers to move in the direction of adding credit and default risk modeling to the IRR analysis process and to the ALCO decision-making agenda.

Case Study: Expanding the ALM Modeling Effort to Incorporate Problem Loans

Discount on a Retail Shopping Center Problem Loan Note

A common theme today is the selling of mortgage notes by banks that are in financial trouble or need to remove loans from their financial statements due to regulatory pressures. These deals are often opportunistic for banks, depending on market conditions.

Note-buyers will buy these mortgage notes at 40 cents in the dollar, often without alerting the business or investor who owns the property. For example, the value of a retail shopping plaza was appraised at US$3.6 million, and the owner borrowed US$2.80 million to purchase the center. The loan interest rate was fixed at 6%. The note was based on a 20-year amortization period but had a balloon payment due in five years of US$2,389,074.67. The US$2.80 million 6% mortgage note was sold to a note-buyer for US$1.12 million (40% of the borrower’s original debt) with 24 months remaining on the note’s original five-year maturity. When the note was sold, the outstanding loan balance was US$2,561,620.47 based on the original amortization schedule. However, the borrower had not made the four most recent payments, so the actual balance outstanding at the sale date was US$2,590,269.34 without considering any late fees or other charges. The bank also had to reverse four months of accrued interest at the time the note was placed on nonaccrual.

Over the three-year period that the loan was outstanding, the bank had accrued interest of US$483,782.99. However, since the borrower had not made payments for four months prior to the note sale, US$51,631.41 of this interest amount had to be reversed when the sale was completed.

When the note was sold, the bank charged off US$1,470,269.34, or the difference between the note’s actual outstanding balance of US$2,590,269.34 and the note buyer’s offer of US$1,120,000. Of course, they had also reversed US$51,631.41 of interest income.

In the typical ALM modeling process, IRR assumptions include the regular payment stream of US$20,060.07 per month and the balloon payment in five years of US$2,389,074.67 for this amortizing loan. However, when the bank decided to sell the note at a substantial loss, the IRR assumptions had to be modified to avoid overstating future cash flows.

The impact of the note sale on IRR assumptions is set out below. All amounts are US dollars.

Original loan

  • Monthly cash flow to reinvest = $20,060.07 for five years = $1,203,604.20.

  • Balloon payment due in five years available for reinvestment = $2,389,074.67.

  • Total cash flows available for reinvestment over the life of the loan = $3,592,678.87.

  • Maturing rate for cash flows = 6%.

  • Reinvestment rate for cash flows: this is set in the ALM modeling process.

Problem loan

  • Monthly cash flow to reinvest = $20,060.07 for three years = $722,162.52.

  • Balloon payment from note sale available for reinvestment = $1,120,000.00.

  • Total cash flows available for reinvestment over the life of the loan = $1,842,162.52.

  • Contractual cash flows not available for reinvestment = $1,750,516.35.

  • Maturing rate for cash flows = 6%.

  • Reinvestment rate for cash flows: this is set in the ALM modeling process.

The “Contractual cash flows not available for reinvestment” in the problem loan section represents the difference in total cash flows available for reinvestment over the life of the loan between the original loan and the problem loan. In practice, the contractual cash flows not available for reinvestment would be discounted to their present value at the contractual “maturing” interest rate.

As seen in the above example, problem loans can distort ALM modeling results unless problem loan assumptions are included in the IRR analysis process.

Back to Table of contents

Further reading

Credit Risk Models and Research

  • Baer, Tobias, Venkata Krishna Kishore, and Akbar N. Sheriff. “Best practices for estimating credit economic capital.” McKinsey working papers on risk no. 11. April 2009. Online at: [PDF].
  • CreditManager (based on the CreditMetrics methodology developed by the RiskMetrics Group and formerly JP Morgan):
  • CreditPortfolioView (from McKinsey):
  • Credit Suisse First Boston. “CreditRisk+: A credit risk management framework.” 1997. Online at:
  • Martin, Richard. “Credit portfolio modeling handbook.” Credit Suisse First Boston, October 29, 2004. Online at: [PDF].
  • Portfolio Manager (from KMV):
  • Task Force of the Market Operations Committee of the European System of Central Banks. “The use of portfolio credit risk models in central banks.” Occasional paper series no. 64. European Central Bank, July 2007. Online at:

ALM: Credit Risk Research Studies

  • Jobst, Norbert J., Gautam Mitra, and Stavros A. Zenios. “Dynamic asset (and liability) management under market and credit risk.” Brunel University, Centre for the Analysis of Risk and Optimisation Modelling Applications (CARISMA) technical report. June 2003. Online at: [PDF].
  • Jobst, Norbert J., and Stavros A. Zenios. “Extending credit risk (pricing) models for the simulation of portfolios of interest rate and credit risk sensitive securities.” Wharton Financial Institutions Center working paper 01-25. July 2001. Online at:

ALM Modeling and Training Companies

Back to top

Share this page

  • Facebook
  • Twitter
  • LinkedIn
  • Bookmark and Share