More than five months after the Home Affordable Refinance Program (HARP) 2.0 was announced and there is not much seen in terms of the program's impact on prepayment data, according to FTN Financial in a report released yesterday.
However, analysts said that there is recent market talk about an uptick in HARP prepayments. In the report, they categorized the size and nature of the risk and offered some thoughts on the possible prepayment projections.
The best way to describe the market’s response so far to HARP 2.0 is underwhelming, they said. Data demonstrated this quite clearly.
The January prepayment report, where many market players had believed HARP 2.0 would start to show a large effect on prepayments, was actually lower versus the December print for all the in-the-money gross weighted average coupons, according to FTN analysts.
Additionally, the slight rise in the refinancing response for the February and March data sets is partly shown by the fact that primary mortgage rates were slightly less in January and February verus November and December, according to FTN analysis.
Anecdotal evidence has demonstrated that some HARP 2.0 activity has started, but this is hardly the response that most analysts – and prepayment models – expected when the program was
initially announced last fall.
However, some equity analysts covering the large banks and REITs are starting to warn participants about HARP 2.0 as a 2Q12 phenomenon as these institutions continue to ramp up capacity for the program.
The two biggest mortgage originators and those who have the most potential HARP volume are Wells Fargo and Bank of America, analysts said.
They cited an April 13 investor packet where Wells noted that 15% of originations in 1Q12 were from HARP and that 76% of all mortgage applications were refinancings, which is very similar to 3Q11 and 4Q11.
However, analysts reported that Wells normally aggressively solicits refinancings for high-quality borrowers in its course of business. Thus there is not much to these disclosures about possible future actions.
FTN stated that the five largest conventional fixed-rate 30-year MBS originators, according to outstanding balance from largest to smallest, are: Wells Fargo, Bank of America/Countrywide, JPMorgan Chase, Citigroup and Ally Financial/GMAC.
Of the top five large servicers, Wells and BofA represent the bookends of potential HARP risk: Wells has the lowest and BofA has the highest.
The potential HARP risk at Citi and Ally/GMAC are 20% and 21%, respectively, FTN analysts reported. Meanwhile, Chase is tied with Wells at 13%.
Furthermore, FTN noted that Wells has by far the highest percentage in the “willing and able” category at 56%. This is key they said as it relates to HARP. Mortgage banking at Wells in 1Q12 was up $506 million, or 21% quarter-over-quarter. This is 8% higher origination volume with higher fees and with only 15% of the business coming from HARP. FTN analysts concluded based on these facts that Wells does not need HARP at this point to run its mortgage business well.
Therefore, analysts still think that the biggest possible risk is from BofA and, to a lesser degree, Citi and Ally/GMAC.
If the models are to be believed, there is a fair amount of price risk in 5.0s through 6.0s in Wells pools if the bank actively pursues HARP 2.0, FTN stated. Analysts are not sure that will happen. However, it is important for investors to note what the potential risks are.
They summarized that buyers should look at the following points about HARP 2.0: Prepayment speeds in 1Q12 have not been in line with original estimates and should move at least somewhat higher for the target HARP cohorts. Additionally, the target HARP cohorts are still the 2005-2008 vintage conventional 30-year 5.0s through 6.5s. Those outside of this target should not experience a lot of HARP exposure, analysts stated.
For buy-and-hold investors, analysts suggested that there are many opportunities now to express a view on carry. Fast money buyers will probably look for better entry points as the prepayment landscape becomes more clear.
