Last quarter, in response to the well-publicized regulatory guidance highlighted in SR Letter 07-01, Interagency Guidance on Concentrations in Commercial Real Estate, we introduced Part I of a three-part series on commercial real estate (CRE) portfolio stress testing for community banks.1 Part I introduced the benefits of stress testing and emphasized the need for institutions to quantify the impact of changes in economic conditions, asset quality, earnings, and capital. A multi-phase approach was suggested, grounded by strong management information systems and CRE portfolio segmentation capabilities.
This second part discusses a number of suggestions for testing specific segments of the CRE portfolio and analyzing the results. At this point in the process, we assume that loan portfolio segmentation and aggregation capabilities have been established.
Stress Testing Individual Sectors of the CRE Portfolio
The most vulnerable sectors in today's market are likely to be commercial loans tied to residential housing development activities (i.e., land acquisition and development, condominium construction/conversion, and speculative residential construction financing). However, given other emerging market conditions, it may be prudent to also test other loan sectors that may be affected in the future. Increasing energy costs and unemployment rates, lower consumer spending rates, or inflationary considerations, for instance, may affect other sectors which have received minor attention until this point in time.
For our purposes, the most critical factors that can be tested are those that increase the probability of loss by directly impacting property cash flows and underlying asset values. Figure 1 provides examples of shock factors that should be considered for CRE portfolio sectors.
The decision to stress test a particular sector will be highly dependent upon the size of a portfolio relative to other sectors and the specific geographic and market trends affecting that sector. Decisions should not be based on the strength of underwriting and loan structuring practices for that sector, the reason being that conservative underwriting criteria, such as high outgoing DCRs, low LTVs, aggressive hold-backs, cash collateral supplements, and other upfront risk mitigation strategies, should inherently cushion the impact of the results.
Residential land acquisition and development. The most prevalent risk associated with this portion of the CRE portfolio is a decline in demand, which results in longer project sell-out periods or declines in sales prices. For this portfolio sector, the most relevant stress test is to lengthen the absorption rate. Current-, best-, and worst-case scenarios for this factor should be considered, with the current absorption rate supported by documented market data from reliable sources. Using these lengthened absorption rates, determine how long it will take for the project to sell out. Under these stressed conditions, how much inventory will remain for sale at loan maturity? How long will it take for loans within the portfolio to be paid in full under these circumstances?
Many lenders may have a false sense of comfort when there are formal contracted lot sales agreements between the developer and third party builders, and they may be tempted to bypass lengthening the absorption rate. Using a best-case scenario could potentially accommodate these considerations. However, it is still prudent to also develop an absorption downside scenario that anticipates that current market conditions will continue to negatively impact the ability of certain developers to honor their contractual obligations.
No matter how low market prices may be, some areas may continue to experience slow absorption rates. Others may not, depending on how readily the individual market is willing to move to purchase on a decrease in prices. Consequently, a separate scenario or a combination of both a decline in price and a slower absorption rate might make sense. It is important to remember that decreased sales prices will have an impact on the amount of funds available to meet minimum lot release prices contracted under the loan agreement. Depending on the level of price decline, lower than anticipated principal pay downs could result in a longer term to payout and/or residual loan exposure at the end of the project sell-out. Institutions may want to consider a worst-case liquidation price scenario for loans that fund projects in highly depressed markets or loans that are risk rated watch and/or classified. Institutions may also need to consider diluting net sales proceeds for increased marketing costs in order to move lots or for liquidation services.
In addition to affecting the rate and level of project cash flow and collateral values, lengthened sell-out rates or lower than anticipated unit prices can also translate into inadequate interest reserves. Institutions that permit bank-funded interest reserves (or finite borrower-funded interest reserve escrows) should test the adequacy of the reserves in the face of a protracted market downturn to determine how much in scheduled interest payments is at risk for the portfolio. What alternative sources of funding are available to carry the project without extending additional funds to replenish the reserve when the sell-out is delayed for a substantial period of time? For loans that are priced at variable or adjustable rates, institutions should also factor in the impact of increased interest rates on the adequacy of interest reserves.
An institution with a loan portfolio that contains a large number of loans funding projects in mid-process (i.e., preliminary development or infrastructure completion) might also want to consider stress testing this portion of the portfolio to determine the impact of increased costs on the portfolio. Consider the following scenarios:
The same considerations can be applied to residential construction (vertical construction) loan portfolios. Application of stress tests measuring the impact of lengthened absorption rates, a decline in sales prices, a loss of pre-sales, and increased construction costs and interest rates are applicable. All tests should be based on documented market assumptions and should be designed to measure the impact on the primary and secondary sources of repayment.
Testing raw land inventory portions of the loan portfolio should also be considered. If a significant concentration exists, management may want to consider testing the impact of a longer-term debt service carry burden and analyzing a decline in property value on LTVs.
Residential and commercial condominium development and construction. During the last several years, the real estate market has seen a proliferation of ground-up condominium development and condominium conversion projects. Demand for this type of property at scheduled prices has waned in some markets. The same assumptions made for residential land development and construction also apply for this type of project (i.e., absorption rate declines, lower sales prices, increased project carry and interest costs).
Delays in condominium sales often translate into reliance on the rental market to supplement cash flow until market conditions improve. Consequently, stress testing should factor in the potential for a shift in repayment sources from unit sales to rental income. This may prompt institutions to consider including a separate analysis of the impact of these conversions on property cash flows and property values. As more properties are converted to rentals, it may also be prudent to consider the impact of competing supply on the apartment sector.
Multi-family residential. The most common type of stress testing for this portfolio segment is to test for an increase in vacancies or rent concessions. Regional economic considerations, such as the loss of a large regional employer or government or military shrinkage, as well as increased competition from condominium conversions and other factors, are likely to affect the occupancy rates in the apartment sector. These factors should be considered in establishing the downside simulations.
Testing for the impact of increased expense levels for multi-family housing projects may also be important in today's environment due to increased energy costs or other inflationary pressures that affect property costs, especially projects where landlords bear the burden of utility costs.
Loans of this type are often subject to interest rate reset risk. Lenders can stress test for interest rate reset risk by evaluating the portion of the portfolio with reset dates occurring within the next 6 to 18 months to determine whether current and stressed occupancy rates are strong enough to support an upward adjustment.
Leased commercial office. A slowing job growth rate and the impact of economic slowdowns on certain service sectors each have the ability to impact the performance of leased commercial office properties. In applying stress tests for this sector, lenders should consider the dynamics of regional economies when determining the degree to which stress tests should be applied to this portion of the portfolio. Properties with tenants in business lines that are currently vulnerable to economic trends should be evaluated for the impact of increased vacancy rates and/or declining rents at levels supported by regional economic data. Properties supported by short-term lease arrangements should also be tested for increased vacancy levels.
For properties where expense pass-throughs do not apply, expenses should be tested for inflationary increases in costs, particularly for utilities and maintenance. Properties with short-term leases should also include increased marketing costs and tenant improvement costs to support re-leasing activities.
As with all other sectors identified in this article, interest rate sensitivity should also be considered on this portion of the portfolio, particularly for fixed-rate loans subject to interim rate resets or loans tied to floating rate pricing arrangements.
Industrial, warehouse, and distribution. Tenant credit risk is the most pervasive concern for this sector. Therefore, stress testing this portion of the portfolio should include a review and an evaluation of industry concentrations in the portfolio that are most vulnerable to declining economic trends. Increased vacancy rates, lower rental rates, and increased expenses (depending on the lease terms) should be factored into the stress testing scenarios for this sector at levels reflecting conditions of the industrial markets served by the tenant base. Particular attention should be paid to those with tenants whose industries have supported the housing market or other vulnerable segments of the economy. Properties are likely to be single-tenant facilities; therefore, worst-case scenarios should simulate prolonged vacancy periods and/or reduced rental rates and increased marketing costs on property cash flows. LTVs should be recast based on reduced levels of projected net operating income (NOI) as a result of these adjustments.
Leased retail. A weakening economy is likely to continue to impact consumer spending, so portfolios with a high concentration of loans supporting this sector may face challenges, as certain retailers face financial pressures due to lower consumer demand. Stress testing for this portion of the portfolio should include an analysis of the same fundamental factors highlighted in other sectors: increased vacancy rates, lower rental rates, and increased property costs (if leases are not triple net). Adjustments to LTVs should also be recast based on reduced levels of projected NOI.
Analyzing the Results and Developing Conclusions
The next step of the process is to determine what the overall impact to earnings and capital will be if simulated situations become a reality by overlaying the results of the stress tests (see Figure 2). It is important to evaluate how declining DCRs and increasing LTVs resulting from stressed scenarios will impact problem loan levels and nonperforming assets. As problem loan levels rise, management needs to consider how interest income may be negatively impacted. The allowance for loan and lease losses (ALLL) under stressed scenarios should also be reassessed. Ultimately, management should review this potential impact on capital based on best- and worst-case scenarios.
There are a variety of approaches to stress testing a CRE loan portfolio. As we have demonstrated, sophisticated models are not always necessary to define potential areas of risk and to determine the impact on earnings and capital. Focusing on simple factors that may affect debt service capacity and collateral adequacy will provide a satisfactory foundation for assessing an institution's staying power during negative market conditions.
We will conclude our series on stress testing by highlighting the role of the board and management in developing mitigation strategies and contingency plans based on the final outcomes of earnings and capital analysis in Part III of this series. In addition, Part III will provide suggestions for developing policies and procedures to promote the ongoing success of stress testing programs.
The views expressed in this article are those of the author and are not necessarily those of this Reserve Bank or the Federal Reserve System.