After a series of missteps and failures by several specialty finance companies that depended on asset-backed securitization as their primary funding source and "gain-on-sale" accounting, the equity markets have become wary of the quality of earnings reported by these companies.
Some analysts have gone so far as to suggest that forcing all finance companies to account for profit on a "flow" or "portfolio" basis (and perhaps even stop funding their business via securitization) is the only way to accurately represent earnings and restore investor confidence. Clearly there have been problems (in some cases, perhaps even abuses) of gain-on-sale accounting. However, when properly applied it can provide a level of transparency and an emphasis on creating value that traditional flow accounting does not.
Fundamental analysis of any company attempts to measure value by discounting current and future cash flows at a cost of capital. Its foundation is based on the obvious premise that investing in a company should yield a cash-flow return commensurate with the risk. Loan originators such as banks and finance companies are certainly no different in this respect. They raise capital from a variety of sources and invest it in loans that yield a stream of cash in the form of principal and interest payments. In addition, corporate activities such as origination and servicing require cash outlays.
Fundamental valuation of a loan-origination business should reflect the net present value of current and future loan production, less overhead costs. The greatest challenge in valuing these businesses is forecasting the cash flows that accurately reflect the loans' anticipated loss and prepayment performance. Though obviously difficult, valuing loan originators cannot be accomplished without trying.
That being said, let's review the basics of gain-on-sale accounting. For those asset sales subject to "gain" accounting, the company books a net profit based on the cash received on the sale transaction, plus a valuation of projected residual cash flows retained, minus the capital the company advanced to place the loans.
The controversy has been about the "assumptions" -- the projected future patterns of prepayment, default, and loss severity -- that must be made to value the residuals.
Clearly, projecting the future is at best a difficult task and usually proves inexact. However, if done honestly and with sound analyses, the projections will reflect the company's best and most informed view of the expected performance of its loan pools. It is unproductive to criticize these firms just because their projections prove to be inexact. Rather, we should judge the realism of their projections and recognize that changing economic conditions will drive honest differences.
Any accounting method should be viewed and judged by the transparency provided and the information revealed to current and potential investors as they place a value on the company.
As I mentioned above, the fundamental value of a lender is the net present value of current and future loans, less overhead costs. If gain-on-sale accounting is done consistently with the objective of honestly portraying an informed projection of future loan performance, then the present value of current and future gains on sale, the net of overhead costs, is exactly the same as the fundamental value.
Thus gain on sale profits are an appropriate indicator of the contribution to the value of the company of loans that were originated and placed in the pool. In addition, the detailed assumptions provide investors with an informed view of the components of the pool's value.
The fundamental value of traditional portfolio lenders with flow accounting is really no different than securitizers using gain accounting. It is surely obvious that these portfolio lenders are generating value by originating loans, funding them with capital, and servicing them. It is also surely obvious that the ultimate cash flows of a portfolio lender will come from the same loan payment sources -- the net of future prepayments, defaults, and recoveries -- as with those firms selling assets subject to gain accounting. Therefore, the ultimate investor criterion for evaluating lenders is the same.
So, how does flow stack up against gain accounting in providing transparency and valuation support? The answer is: poorly, given that with flow accounting all of the current information regarding loan quality is typically revealed in two gross numbers, the net additions to and subtractions from loan-loss reserves in the current reporting period.
These reported adjustments to loan-loss reserves, in aggregate, reflect the anticipated and actual performance of all existing and new loans without adjusting for portfolio seasoning.
So how is an outside observer to translate these gross numbers into an understanding of the quality of new loans originated? How does an analyst determine whether existing loans are actually performing in line with the expectations management had at the time they were originated?
In fact, many lenders employ static-pool analysis to anticipate future performance of new loans. These analyses result in projections of loan performance based on historical experience. In essence, this comes very close to the projections embodied in gain-on-sale accounting.
The weakness is that portfolio lenders do not typically reveal these assumptions to investors. Thus, though the projections are still being made, they fail to provide the level of information to investors that the publicized projections in gain-on-sale accounting afford. As an aside, many securitizers who use gain-on-sale accounting apply static-pool analysis to arrive at their projections. The potential flaw in this is that typical static-pool analysis does not account for past or future changes in regional and national unemployment rates, housing price inflation, interest rates for competitors' (and substitute products) and other factors known to affect loan performance.
In addition, static-pool analysis is often flawed by assuming that loan pools are composed of loans with identical or similar characteristics, thus similar performance. However, it is a rare lender who never changes products, underwriting, or servicing practices. Plotting curves of future default and prepayment rates based on historic performance will certainly result in errors, as we have seen in the recent past.
The real problem facing lenders using gain accounting is that they reveal so much information that they are subject to criticism more often, and on more fronts. Is it better that portfolio lenders have the luxury of not being held to projections? Furthermore, because of the aggregate nature of flow or portfolio accounting information, portfolio performance (and product design and pricing) problems may be masked or unrecognized for a long time -- sometimes several years. By the time quality problems are revealed, the costs to investors are much higher, as we saw in the 1980s with some savings and loans.
In the short run, securitizers that apply gain-on-sale accounting have the power to fool the markets (and themselves) by booking illusory profits based on error-plagued and unrealistic assumptions and projections about future losses and prepayments. But, as we have seen recently in the specialty finance industry, investors can quickly identify and punish these companies. When their investors withheld capital, several of these companies failed and are no longer in business. However, flow accounting is also susceptible to errors in judgement, and even manipulation. And, as we have seen with savings and loans -- and are likely to see with some banks when the economic cycle turns, if history is a guide -- they can fool the market into believing that they are profitable, at least for a while.
In the end, the successful companies will be those that are honest and forthright and maximize value, regardless of their accounting methods. The economic consequences of gain and portfolio accounting are identical.
If gain companies can avoid the obvious missteps of the past (that is not to say that there will not be honest downside surprises) and keep investors up to date on performance, their reputations will grow.
Investors' confidence in them may even grow to exceed that for "portfolio" lenders with less transparent accounting methods.
The advantages of more transparent reporting and a focus on managing the value of loan portfolios (rather than the appearance of earnings) that are possible with gain-on-sale accounting have been lost in the recent controversy. Suggesting that flow or portfolio accounting improves investors' ability to understand a lender's value ignores the obvious: How could less information possibly be better?