The financial press ran a number of stories last week that referred to a Federal Reserve study on the economics of the mortgage market. The study, entitled Credit Scoring and Mortgage Securitization: Implications for Mortgage Rates and Credit Availability, raised the possibility that mortgage securitization, and by implication the existence of Fannie Mae and Freddie Mac, does not necessarily lead to lower mortgage rates.
The main conclusions of the study are that conditions exist where securitization may fail to lower the mortgage rate. Additionally, the authors conclude that a decline in the mortgage rate causes increased securitization rather than the other way around. In our opinion, the study is undertaken using highly controlled conditions and assumptions, and does not fully reflect the dynamics of the rate-setting processes. Also, the main conclusions of the study are based upon the credit risk of mortgages and do not account for the effect of prepayments on both issuance volume and borrower behavior.
Our interest in this study stems from the fact there is an undue amount of attention being generated on the conclusions of the study without an equal amount of attention being devoted to the methodology and realism of the assumptions used to derive the conclusions. At the outset, we would suggest that it is difficult to determine causation between securitization and mortgage rates. Yes, lower rates lead to increased securitization because of increased refinancing activity. Also, higher rates may also lead to increased securitization. How else would we explain recent activity - mainly cash out refis and new home purchase in the midst of bearish market conditions?
The Basis (or Lack Thereof) of the Argument
Without getting into the historical details and at the same time without taking a stance on the issue, the topic of GSEs and related issues of implicit guarantees, mission creep, political maneuvering, debates and congressional hearings has been in the media limelight. Our main concern with this study is that it adds to the drama without necessarily providing the underlying basis for the conclusions.
With respect to the main conclusion that the presence of securitizers (by reference the GSEs) does not necessarily lower the mortgage rate to consumers, our thoughts are as follows.
One of the key arguments is the modeling of the rate-setting process by the securitizer. Contrary to the argument in the paper, the securitizer is not a monopolist in the economic sense of being able to control the interest rate at which the loans are bought. In our opinion, the pricing of mortgage loans is determined more by MBS levels and spreads than by any unilateral determination by securitizers. MBS spreads are determined by a complex interaction of various factors, such as supply, agency portfolio activity, cross market supply and performance, prepayments, volatility, the CMO bid and the like. For instance, during periods of premium price compression, where the market discounts high coupons, originators may choose to hold more excess servicing in the MBS execution.
To the extent that the rate set by the securitizer is market determined and hence has an embedded liquidity premium, the originator is provided with the mechanism of passing through some or all of the liquidity premium to the borrower.
Given that the market for origination of loans is highly competitive in all channels, it is difficult to envision situations where most of this premium is not passed on to the consumer. In some respects, the securitizer, by creating credit and loan attribute standards, contributes to the creation of this highly competitive market place.
The transfer of the liquidity premium to the consumer takes on an additional element of importance during a period of high refinancing. For originators, who are also servicers of the loans, refinancing-related prepayments lead to impairment of the capitalized servicing asset. During such periods, the originator has the additional incentive to pass on the liquidity premium to consumers with the purpose of capturing business and generating fees to counterbalance impairment losses. Therefore, the exclusion of the effect of prepayments from the study leads to incomplete conclusions.
The U.S. mortgage market has been characterized by multiple innovations, where the securitizers have continually changed documentation standards to incorporate additional loans into the "conforming" fold. History is replete with instances where the pricing on such loans has improved due to the fact there is a securitized bid for such loans. To the extent that this translates into lower pricing for consumers, it is difficult to argue that the securitization-related liquidity premium is not passed on to consumers. Additionally, the effect of such "benchmarking" also spills over to the "non-conforming" markets, which typically trade "behind" agencies; the very presence of a benchmark thereby improves pricing on such loans. In fact, the best point estimate of this benefit may be derived every year as the "all-in" conforming to non-conforming jumbo loan spread. For instance, assume the conforming limit is $252,700 and conforming loans are priced at 8% while jumbos are at 8.25% (which, at this writing, are current 0-point rates). If the conforming limit is raised to $270,000, a loan amount of $260,000 that was priced at 8.25% is now priced at 8%. This is perhaps, the best example of how the liquidity premium is passed on to the consumer.
While others may not agree with our thoughts, it is our belief that the most important aspect of the study will be its political impact, specifically the impact of its headline adding further to the GSE debate. While there are individuals and entities a lot more qualified than us to opine on both sides of this debate, our motivation here is not to implicitly or explicitly align ourselves with the issue. Rather, as students of the markets our motivation is to indicate that this study, which has significant headline value, has been undertaken under fairly controlled conditions without completely accounting for either the market's dynamics or a realistic assessment of the issues involved.