Publications

Introduction

This paper gives a brief introduction to cost-of-funds, (also known as transfer pricing) and the difficulties of setting the cost of funds for commercial real estate (CRE) lending operations. It then suggests a relatively straightforward approach which avoids these difficulties.

The Importance and Difficulty of Setting the Cost-of-Funds

The Cost-of-Funds (CoF) is the rate charged by the institution's central treasury on money they give to individual lending units when those units request funds to lend on to customers. The CoF is the primary cost used in measuring a business unit's profitability and typically it directly impacts the compensation of the lending teams. It therefore acts as a strong incentive in defining the nature of the assets being originated. Very simply:

The CoF has four potential components:

• The risk-free¹ rate for the term and rate profile of the loan to the customer
• An operating cost
• A subsidy from senior management to guide the relative growth of businesses
• The cost of risk

This paper is focused on the cost of risk. A common practice is to link the cost of risk to the capital² to be held for the asset. Pricing according to capital has several problems and is particularly difficult for CRE as CRE loans have long-terms and are highly-structured with time-varying risk profiles. Pricing according to the regulatory capital is especially problematic because regulatory capital typically only reflects the average risk of similar assets. The objective of regulatory capital is to ensure that the portfolio as a whole is backed by sufficient capital so regulatory capital does not necessarily reflect the risk of individual assets, and typically does not capture the changing risk profile of CRE assets over time. Regulatory decisions to require additional safety may also mean that the CRE assets are over-capitalized compared with other asset classes³. The net result is that the business units are incentivized to lend in such a way that other institutions can arbitrage against them. The arbitrage results in the lenders taking on more risk than the income from the customer would justify, as explained in our paper Capital, Arbitrage and CRE Lending.

Pricing according to Economic Capital (e.g., RAROEC⁴) has the advantage that it can account for individual deal characteristics and therefore differentiate between alternative deal structures, but the long-term, time-varying risk profiles of CRE assets means that the implementation of RAROC for CRE assets is highly problematic, as explained in Enterprise Risk Management and CRE Lending.

As an alternative to capital, another common practice is to set the CoF according to benchmarks which are considered to be similar to the loans, e.g., setting the CoF according to the rate on bonds of the same rating as the loan. The problem then becomes defining what is 'similar' because seemingly small differences in the CRE deal structure can have a large effect on the risk. Consider the risk estimates listed in Table 1.

Table 1

Each of these deals has the same LTV and DSCR and is identical to the Base Case other than one variation in the property or financing structure. Notice as an example that adding a sweep covenant (Variation 1) has a very different effect than adding a sweep reserve (Variation 6) although many people judging them qualitatively would argue that the effect on risk is expected to be similar. The full description of this analysis is given in Beyond the Simplicity of DSC and LTV.

Under many regulatory capital regimes all of the deals in Table 1 would be accorded the same capital, and pricing relative to the regulatory capital would not recognize the wide variation in risk for the individual deals. Similarly if the selection of benchmarks did not take into account the detailed deal features, the incentives for deal selection, structuring and pricing would be skewed.

Recommendations

In setting the Cost of Funds for CRE loans, we believe the following approaches should be used:

• Set the cost of risk using NPV of loss⁵ to capture the time-varying risk profile of CRE loans. An alternative to using the raw NPV result is to also calculate the NPV of loss for a series of benchmark bonds, and then take the market pricing from the bond with the closest EL%⁶.
• To estimate the expected loss per year and Beta⁷ use a risk model which takes into account as many deal features as possible. There will always be special features for each deal that need to be judged qualitatively by experts, but to the extent possible, quantify the known features into a model to focus the debate on what is truly special about each deal.
• Centrally set the model logic and parameters to reflect the experts' proprietary views on the markets. Some of the parameters such as the average expected forecast for property values are highly subjective and depend on combinations of forecasting models and expert opinion. The risk model then structures, quantifies and prices the implication of those expert views and shows the consequent risk for the loans.
• The risk estimates should ideally show the profile of the sources of risk in an actionable form so that the lending teams can structure around the risk profile created by the underlying properties. Examples of actionable risk profiles are shown in Dissecting CRE Loan Risks - Lease, Tenant, Interest Rate and Refinancing Risk

Conclusions

Regulatory compliance requirements such as capital and stress testing reports are important mechanisms to ensure the safety of institutions and protect the financial system, but even though they strongly influence the capital to be held, they are not designed to control pricing, and they do not need to be used control pricing for the simple reason that the cost of capital can vary institution-by-institution and asset-by-asset⁸. This means that institutions that incentivize their business units by using regulatory capital to guide the cost of funds are using capital in a way that was not intended, and will suffer accordingly. This is especially true for CRE where the complex nature of the assets and their complex time profile widens the gap between capital and the risk of individual deals.

The CRE lending business has been moved to a cross-roads by the new regulations. Because of its complexity, CRE has been one of the last credit businesses to use statistical models in managing the business. The new regulations, directly or indirectly, are forcing capital to be linked to measures of portfolio risk. However, institutions who set their CoF and profitability measurement according to the regulatory capital will actually be taking a step backwards and will be mis-directing their lending businesses. To avoid being arbitraged, institutions need to take a step forward and set their cost of funds according to the best assessment of the risk of each individual loan and structure.

Dr. Chris Marrison
CEO, Risk Integrated
Chris.Marrison@RiskIntegrated.com

1. Strictly speaking it is not the risk-free rate but the rate on long term debt to the bank, e.g., the 'A' or 'BBB' rate.
2. The cost of capital is typically taken to be the additional expected return on capital multiplied by the percentage of the loan balance to be backed by capital. For example if the expected return on capital was 18%, the risk free rate was 3%, and 6% of was the amount of capital to be held against the loan, the cost of capital; would be (18% - 3%) x 6%.
3. Some regulators consider CRE to be an especially risky asset class, or one in which the risks are not well measured and they have responded by requiring higher capital multiples for CRE compared with other assets. This excess capitalization can be compensated for by reducing the expected rate of return on capital for CRE assets, but that just moves the problem to defining the rate of return.
4. Risk adjusted Return On Economic Capital
5. Net Present Value as in the article Enterprise Risk Management and CRE Lending.
6. EL% is NPV of loss divided NPV of balance, i.e., the spread such that the NPV of income equals the NPV of loss.
7. Beta is the correlation term which sets the discount rate in the NPV calculation.
8. For 'the use test' it is possible to use the same models both for capital calculation and calculation of the risk profile, but the results are used differently for each application.