We estimate that the foreclosure potential among the 40,000 marginal borrowers that would have received FHA-insured loans at lower MIP rate, would be significantly higher than the current average FHA foreclosure rate.
Just three weeks before the end of the Obama administration, HUD announced a reduction the annual FHA mortgage insurance premium for high loan-to-value loans of .25%. Then Secretary Julian Castro announced the reduction and in his statement estimated that it would save new FHA-insured homeowners an average of $500 per year (which is equivalent to $42 per month). Secretary Castro cited the health of the FHA insurance fund and the anticipated rise in interest rates as the reason to help creditworthy working families achieve the dream of homeownership. Within hours of President Trump’s swearing in, HUD announced that the premium reduction was “suspended indefinitely.”
Various experts in the housing policy field have offered observations regarding the effect of the suspension. The National Association of Realtors estimates that 30,000 to 40,000 families will now be priced out of the housing market. The co-director of the Housing Finance Policy Center at the Urban Institute reported that foregoing the MIP rate cut strengthens the FHA’s financial situation and allows the FHA Mutual Mortgage Insurance Fund to build its buffer.
Besides lowering the potential risk to the MMI fund, we believe that the suspension of the MIP reduction also protects neighborhoods and families from loss of wealth that results from foreclosure of nearby properties. According to an article from May 2016 published by the Urban Institute, borrowers with lower FICO scores were more likely to disproportionately gravitate towards FHA financing when FHA cut its annual premium by half a percent at the beginning of 2015.
If HUD had not suspended the MIP rate reduction in January 2017, we believe that it would have increased the attractiveness for high risk borrowers. For those borrowers whose financial conditions are so marginal that a very nominal differential in housing costs would make the difference between a performing loan and a non-performing loan, those loans would most certainly be classifiable as high risk. As such, one would expect a higher-than-average default and foreclosure rate, and one could expect an especially high rate of claims against the MMI Fund.
The HUD announcement of the MIP rate reduction alluded to HUD’s internal evaluation of risks, concluding that it was the judgement of the outgoing administration that the risks were sufficiently moderate when compared to the anticipated benefits to the 30,000 to 40,000 families that might be able to afford homes as a result of the rate reduction.
The risk that is unlikely to have been included in the calculation, however, is the collateral damage inflicted on families when foreclosures occur on properties near those of performing borrowers. The financial damage is the loss of equity that performing borrowers may have in their home. The equity in their home is most often their only significant asset. The cost of neighborhood destabilization that results from foreclosures, while difficult to quantify, most certainly exists.
Although there are a range of estimates of the size of the spillover impact of foreclosures on nearby properties, housing economists generally agree that there is a negative spillover effect and that its magnitude is related to the proximity, number and timing of foreclosures. We estimate that the foreclosure potential among the 40,000 marginal borrowers that would have received FHA-insured loans at lower MIP rate, would be significantly higher than the current average FHA foreclosure rate. Based on our internal estimates of average house value, loan size, proximity factors and value impact at various housing densities, we believe that the suspension of the MIP rate reduction preserved hundreds of millions of dollars of homeowner equity for tens of thousands of families. Until the impact of foreclosures on families and neighborhoods is incorporated into the policy calculus, the methods for setting the MIP rate at any given point will be incomplete.