Distressed credit portfolio valuation
The prolonged global economic downturn, begun in 2008-2009, has resulted in large distressed credit portfolios in banks and is putting them in potential non-compliance with Basel III capital requirements. Here’s what bankers should know …
- A combination of economic conditions and banking regulations have created both supply of and demand for distressed credits in the global capital markets.
- The price of distressed credits is generally not the same as their value, and banks must be careful to estimate fair market value.
- Distressed credit experts often have incentives to bias valuation estimates downwards.
- A combination of certainty-equivalent valuation and econometric methods are usually necessary for estimating fair market value of distressed credits.
- Distressed credit sale negotiations depend critically on value estimate accuracy, reliability, and supporting evidence.
- Auditors tend to object to valuations based on non-standard estimation methods, but the methods are usually necessary in the case of distressed credits.
Do not read this article unless you are a banking professional directly involved in distressed credit portfolio transactions. You’ll be sorry otherwise; I promise. :- )
1. The distressed credit market
Bankers I speak with say many commercial banks are sitting on large distressed credit portfolios–much of which have been charged-off for financial and regulatory reporting purposes–and this is starting to cause problems with Basel III risk-weighted capital and other banking regulatory compliance. This causes problems for some and, of course, opportunities for others. It turns out that understanding the problems and opportunities is quite important for understanding how to determine the value of distressed credit portfolios, so I’ll explore these in some detail before turning to more technical valuation issues.
Distressed credit portfolio supply. The most significant factors influencing the supply of distressed credit portfolios to the capital markets is the confluence of economic downturns and banking regulations. As but one example, Brazilian banking regulations effectively require banks to comply with Basel III requirements while at the same time recording credit loss allowances and credit charge-offs in their financial statements based entirely on a credit’s time-in-contractual-default; generally regardless of collateral and guarantee values, etc. This tends to make the actual market value (MV) of distressed credits exceed their net financial reporting value:
MV > credit principal balance – credit loss allowance
Combine this divergence between economic reality and financial/regulatory reporting with Basel III (or similar) capital constraint regulations and the recent Brazilian economic downturn and several unpleasant things happen in sequence: increased credit defaults; increased credit loss provisions and allowances; increased credit charge-offs; decreased regulatory capital; and, finally, decreased regulatory compliance.
At this point, there really is only one thing a bank can do to mitigate (though not solve) the problem: sell distressed credits. Hence the regulatory compliance-induced supply of distressed credit portfolios to the capital markets.
While I might have overlooked a few factors influencing this supply, what I hear from banking industry professionals suggests that this about covers it. So, let’s move on to the demand for distressed credits.
Distressed credit portfolio demand. I would guess the reader, having read thus far without being dissuade by the technical jargon, is quite familiar with the recent phenomenon called the “search for yield” resulting from the zero interest-rate policy monetary environment beginning in 2008. So, rather than waste time by explaining it here, I will simply point out that the demand for distressed credits has increased rapidly.
To get an idea of the nature of this increased demand, I think it’s instructive to take a look at the a Google search by clicking on …
… and note the breadth of the large hedge fund (e.g., KKR, Blackstone) and alternative asset management firms involved in the market.
In short, this means that in the current capital market environment banks with distressed credit portfolios now have opportunities to sell into very large pools of investment funds; primarily because there is substantial demand for distressed credits because higher potential investment yields than are available elsewhere in the capital markets. As we will see, however, capturing the opportunities does require some care if one intends to maximize recovery from distressed credit portfolios.
2. Selling price is not the same thing as value
To fully understand how and why to “value” a distressed credit portfolio, it’s important to be clear about the distinction between value and price. Most basically, in financial economics value is a buyer’s or seller’s subjective, unobservable perception while price is an objective, observable fact resulting from market interactions between buyers and sellers.
Where people tend to get confused on the distinction is that price depends on the subjective value perceptions of both buyers and sellers, and their market interactions (“negotiations”):
Understanding the relationship shown here is critical to understanding how negotiations of sales and purchases of distressed credit portfolios progress towards where the buyer’s and seller’s subjective valuations converge to the objective selling price through what economists call a tâtonnement process (cf. the notion of a Walrasian auction). In essence, the buyer’s and seller’s subjective valuations are revealed during the negotiation process; resulting (or not) in agreement on value and becoming observable in price.
So, what are the implications of the discussion for how and why to value distressed credit portfolios? Quite simply, because the basic problem facing a bank is how to maximize the selling price of a portfolio, the objective is not to “value the portfolio”; again, value is subjective and, therefore, depends on a particular buyer and seller. The actual objective is to estimate likely selling price of the portfolio under a certain set of market conditions, which critically includes a specific–though hypothetical–set of buyers:
Noting the seller would reasonably accept only the highest bid price for the portfolio (duh) and that it generally takes time to identify, inform, and negotiate with, a number of potential buyers, it can be seen that estimated likely selling price depends critically on the length of time and breadth of potential buyers to which the portfolio is exposed.
What I’ve basically done in the discussion here is to outline the rationale for perhaps the most common objective used in valuation practice, market value; defined in the International Valuation Standards as:
Market value — the estimated value for which an asset or liability should exchange on the valuation date between a willing buyer and a willing seller in an arm’s length transaction, after proper marketing and where the parties had each acted knowledgeably, prudently, and without compulsion.
Note that the definition implicitly captures the basic notion of reasonable exposure to potential buyers over a sufficient length of time, as well as of adequate negotiation, resulting in adequate information revelation and agreement on value … leading, of course, to price. The definition above is consistent with what is referred to as fair market value in the USA and elsewhere because it presents conditions under which a reasonable person would conclude an observed market price was set under fair conditions.
Having settled the matter of what we should estimate and developed a basic idea of why we need to do this (more on this later), we can now turn to how.
For simplicity, I will refer to this estimation objective for distressed credit portfolios as fair market value (FMV) to emphasize the relevant market conditions. To conform to standard professional terminology, I will use the verb value to mean estimate likely selling price in what follows.
3. Expert opinion on value is not necessarily FMV
There is a seemingly common sense–but naive–view of estimating FMV of distressed credit portfolios. According to this view, estimating FMV is easy: Call some external distressed credit professionals or potential buyers; bring in a few to look at credit documentation, etc.; and, ask them for their professional, unbiased opinions about what they would pay for the portfolio.
What could go wrong, you ask? Well, a lot actually: Most distressed credit professionals and buyers are generally experts only on what the portfolios sell for under forced sale conditions; usually forced by regulators requiring a bank to improve capital ratios and liquidity. This is a problem from several perspectives. First, it’s not exactly easy value a distressed credit portfolio under fair market conditions, so many banks generally do not know the extent to which such forced sale valuations approximate FMV. Second, and it should be obvious, forced sale selling price is almost always lower than FMV and if the only valuation estimate is on a forced sale basis, then a bank’s executives tend to agree to sell near this price because they usually have no other reliable information. Finally, and most importantly, it can be shown (i.e., trust me) that such external distressed credit professionals have very strong reputational and financial incentives to under-value distressed credit portfolios: they benefit in a number of unobservable ways by “being conservative”, which is actually true of valuation professionals in general; a good topic for future articles.
So, in short, holding the naive view leading to hiring a distressed credit expert to give an independent professional opinion is a Bad Idea unless the assignment requires the professional to estimate FMV and provide evidence supporting the FMV estimate. Unfortunately, most distressed credit experts are not particularly familiar with FMV estimation methods or how to support the estimates with evidence.
4. Certainty-equivalent valuation methods
FMV depends on risk sensitivities. At some level we all know the value of any financial asset equals the discounted present value of expected future cash flows. But that’s not quite specific enough; cash flow risks and discount rates are necessarily intimately related:
The risk-adjusted expected rate of return R(x), and so FMV, depends on (i) the financial asset’s risk exposures, x, influencing expected future cash flows CF(x) and (ii) the sensitivity of the cash flows to the risks. Specifically, R(x) is an estimate of a fair market’s expected rate of return conditional on the specific risk exposures and sensitivities of the financial asset, based on risk-return characteristics of alternative investments available in (fair) capital markets. So, it represents the opportunity cost of capital with respect to the financial asset because the asset’s risk-return characteristics can be replicated by investing in a combination of alternatively available assets (cf. Ross, 1976, “The arbitrage theory of capital asset pricing,” Journal of Economic Theory 13, pp. 341-360).
Identifying and estimating risk exposures and sensitivities tends to be the main problem when valuing distressed credit portfolios; particularly because there is usually a lot of statistical noise surrounding observable risk factors influencing distressed credit recovery cash flows. Why this is so is a topic for another time but, basically, such noise makes it difficult to reliably measure risk factor sensitivities. Statistical noise tends to bias risk sensitivity estimates towards zero.
So, because accurate, reliable FMV estimates depend on accurate reliable estimates of R(x) and, hence, risk factor sensitivities, distressed credit portfolio valuation using the standard discounted cash flow approach tends to result in unreliable estimates because–among other problems–it generally fails to account for what economists refer to as model risk. The greater the model risk, the lower the reliability, and, so, the lower the estimated FMV. I could expand on this idea, but hopefully this is sufficient for our purposes here.
Certainty-equivalent (risk-neutral) valuation. Fischer Black, co-developer of the Black-Scholes option pricing model, in his 1988 article “A simple discounting rule” (Financial Management 17, pp. 7-11) makes the insightful point that it’s easier to estimate the conditional expectation of a variable than an unconditional expectation (it’s generally a more reliable estimate), and presents a simple method of applying what is known as risk-neutral pricing theory. Setting aside details, an implication of the theory is that if there are to be no arbitrage opportunities in capital asset markets then …
… where certainty-equivalent valuation is a more common term used by professionals for what academic economists refer to as risk-neutral pricing. For our purposes, the most important aspects of the above equation is that (i) a certainty-equivalent level of expected risky cash flow exists for which (ii) the appropriate fair market discount rate is an immediately observable risk-free market rate of return (e.g., the published 1 year maturity U.S. Treasury Security rate):
Because the certainty-equivalent approach does not require both forecasting cash flows conditional on risk factors and estimating the related (fair market) risk-adjusted expected rate of return, substantially more time and attention can be devoted toward understanding, investigating, and estimating the effect of risk factors influencing the distressed credit portfolio recovery cash flows. With such understanding and estimates, one can simply (roughly speaking) focus on deriving accurate, reliable certainty-equivalent cash flow forecasts as I discuss next.
Certainty-equivalent cash flows estimation. As shown, FMV depends on either expected future cash flows or the certainty-equivalent level of the future cash flows. Both require estimation of the relationships between risk factors and cash flows, and the only meaningful way to estimate the relationships (that I’m aware of) is to use econometric methods; a branch of mathematical statistics. Based on reviews of a bank’s credit management systems and interviews with credit management professionals, econometric models similar in structure can developed and estimated using historical data and econometric methods:
Expected credit recovery cash flows can be obtained using expected risk factor realization levels as shown, however what is needed is the certainty-equivalent levels of cash flows and–therefore–of the risk factors as well.
Because we are interested in estimating FMV, we are interested in what traders in a fair market likely believe to be certainty-equivalent levels of risk factors and cash flows. Using the method presented by Fischer Black in his 1988 article, we can again use econometric methods to estimate the relationships between the excess of asset market returns over risk-free returns and the risk factors. With such estimates we can easily derive the certainty-equivalent risk factor levels, most easily understood by considering the following graph:
Certainty-equivalent cash flow levels can then be obtained from the estimated parameter values of the cash flow and risk factor model as …
Certainty-equivalent valuation model. Given the above estimates, the certainty-equivalent valuation model for the distressed credit portfolio has the basic form of a discounted cash flow model; the only difference being how the variables are measured:
Perhaps the most important aspect of the certainty-equivalent valuation model is that it makes (quite) explicit the relationships between risks, cash flows, and discount rates. This has important implications as I discuss next.
5. Bid-ask spreads, negotiation, and information
In my experience, many people think results of sale negotiations are based mainly on interpersonal skills of the negotiators. Maybe this is true in general but when it comes to negotiating the sale of large credit portfolios (distressed or not), information about FMV turns out to be quite important. Consider that, in most all cases, a buyer’s initial bid price is below the seller’s initial ask price, with the final selling price falling between those two values. What causes the final selling price to be closer to the asking price than to the bid price? Negotiation is part of the answer, but information related to the FMV estimate (including supporting evidence) is generally the primary factor influencing where the selling price falls on this continuum:
In my experience, initial bid prices for distressed credits are approximately equal to forced sale liquidation values. One reason for this is that, to date, many banks really do not know the likely FMV of distressed credits and tend to under-value and -price them accordingly. Buyers of distressed credits generally have a clearer idea of the FMV and set low initial bid prices simply to minimize price and maximize return. So, whoever has the superior FMV information–buyer or seller–tends to drive the price towards their preferred outcome:
Accurate, reliable FMV information actually has a number of related benefits to sellers of distressed credits, but for purposes of the discussion here this is the main benefit: it maximizes selling price of the distressed credit portfolio. Why this is so is perhaps the topic for another article, but I will leave it here for now.
6. And the all-important final question
Any banker that has had both the interest and stamina to follow the article this far perhaps has one final, all-important question: Is the certainty-equivalent valuation method acceptable to auditors (and banking regulators)? Although auditors and, to some extent, banking regulators prefer standard valuation methods simply because they are easier to understand and audit/examine, experience suggests that any clear, well-documented valuation report supported by well-accepted theory and statistical evidence is acceptable to them.
As evidence, I can state that I’ve successfully defended many valuation reports based on certainty-equivalent valuation methods including those for distressed credit portfolios against aggressive challenges from auditors from three of the Big 4 audit firms; and that makes for a 100% successful defense rate.
Based on the theory and evidence I’ve outlined above, I think it’s quite safe to conclude certainty-equivalent valuation in combination with econometric methods are generally necessary for estimating FMV of distressed credit portfolios. Moreover, the methods have a number of benefits that make them optimal in many or most cases where banks need to clearly understand and support with statistical evidence the value of their distressed credit portfolios.
Caveats. Please note: (i) views presented above are my own and do not reflect those of others; (ii) like anyone, I’m not infallible and am responsible for any errors; (iii) I greatly appreciate being informed of any significant errors in facts, logic, or inferences and am happy to give credit to anyone doing so; (iv) the above article is subject to revision and correction; and, (v) the article cannot be construed as investment or financial advice and is intended merely for educational purposes. MMc