Yves here. Mortgage giants Fannie Mae and Freddie Mac are implementing changes to their fee structure starting May 1, which roughly raises fees for high-credit borrowers and lowers them for low-rated borrowers. weaker credit. But like Rajiv Sethi, this simplification isn’t entirely accurate (some high-credit borrowers fare better under the new rules) and some of the comments are so wrong they seem misleading, as opposed to ignorant.
By Rajiv Sethi, Professor of Economics at Barnard College, Columbia University and External Professor at the Santa Fe Institute. Originally posted on Imperfect information
Mortgage originators in the United States typically sell their loans in the secondary market, where the primary buyers are government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac. These agencies levy surcharges called Loan Level Price Adjustments (LLPA), which affect the eventual prices paid by homebuyers. Surcharges depend on a number of factors, including the borrower’s credit rating and mortgage loan-to-value ratio. Generally speaking, lower credit scores and higher loan-to-value ratios correspond to riskier mortgages, which have historically been associated with higher fees.
A change to the fee structure due to take effect on May 1 has drawn a lot of heated commentary, mixing factual reports with some exaggeration, distortion and outright lying. According to a report, an (anonymous) mortgage company executive complains that “we need to teach borrowers to baden their credit before applying for a mortgage in order to get the best price.” Advice on exactly how to do it started to appear on line.
If it were really possible to get a cheaper loan by sabotaging one’s own credit rating, the policy changes would indeed be absurd. In the language of the mechanism’s design, the new fee structure would not be compatible with the incentives, and all sorts of chaos would ensue. Consultants (and charlatans) were stepping into the breach to profit from advising potential borrowers on the quickest, cheapest, and most easily reversible ways to appear less creditworthy.
But it’s not what the changes entail. Having a good credit rating continues to confer an advantage under the new fee structure, although to a lesser degree at certain levels of the loan-to-value ratio. In other words, fees have been reduced for borrowers with lower credit scores and increased for those with higher scores, in a way that compresses the difference without erasing or reversing the prior advantage.
That said, there are some confusing changes in the fee structure when it comes to loan-to-value ratios, and it’s worth taking a closer look. First here are the current loan-level pricing adjustments, which expire at the end of the month:
Note that in each column, the rates increase as one moves down to lower credit ratings. And in each row, the rates increase as one moves towards higher loan-to-value ratios (corresponding to lower down payments relative to the value of the property). In other words, riskier loans are more expensive. The only exceptions are at the 80% loan-to-value threshold, corresponding to a down payment of 20%. Here, the fee decreases slightly as one moves to the adjacent column on the right, for those with a credit score of 680 or above. This likely reflects the fact that those who make down payments below 20% are required to take out private insurance, making the loan less risky for lenders. This is not necessarily a violation of incentive compatibility, although one can imagine cases in which it is advantageous to offer a down payment slightly below the threshold. new fee structurewhich will come into effect next month:
Note that credit scores below 639 have been grouped into one row, and those above 740 have been split into three. Additionally, LTVs at 60% and below now appear in two separate columns instead of one. This makes direct visual comparisons a little difficult, but one thing is clear: at no level of the loan-to-value ratio does someone with a lower credit score pay less than someone with a higher score. In other words, you cannot win by sabotaging your own credit rating.
In general, there are two types of borrowers who stand to gain under the new fee structure: those with low credit scores and those with low down payments. In fact, those at the top of the credit score distribution earn under the new structure if they have down payments of less than 5% of the value of the house. The following table shows gains and losses compared to the old structure, with low fees in green and high fees in red (a comparable color-coded chart with somewhat different cells can be found here):
What could be the reason for rewarding those who make particularly low down payments? The goal may be to make housing more affordable for people who have not accumulated significant savings or inherited wealth. But there will be an unintended consequence, as those with reasonably high credit scores and substantial wealth choose to strategically lower their down payments in order to benefit from lower fees. They can do this by simply borrowing more for a given property or buying more expensive properties relative to their accumulated savings. The incentive to do so will be greatest for those hardest hit by the changes, with credit scores between 720 and 760 and down payments between 15% and 20%.
I wonder if the designers of the new policy have taken this strategic effect into account, which will tend to hollow out certain cells in the middle and populate those on the right.
Given all this, how do we assess the changes? Despite some provocations and too simplistic securities, the merits of the new fee structure are debatable and our assessment should depend on the default risks associated with each loan category as well as the overall objectives of the program itself. After all, the reason GSEs play a role in the mortgage market has always had to do with creating conditions for borrowers who would otherwise struggle to get loans. Such initiatives inevitably involve some degree of cross-subsidization. For example, our health insurance taxes are not tied to our health status, our life expectancy, or the likelihood that we will need particularly expensive care in future years. Complaints about the new fee structure seem to implicitly assume that the structure it replaces was fully aligned with program objectives. But that in itself is a debatable proposition.
Some clarity from the administration regarding not only their goals, but precisely how changes in the fee structure should achieve those goals without harmful side effects would be most welcome at this stage. Otherwise, a policy change that may well have some merit will very easily be distorted and demonized.