Housing Market and Mortgage Conditions

To enhance understanding of conditions in the mortgage and housing markets in the Ninth Federal Reserve District and nationwide, we offer visual representations of Mortgage Originations, Mortgage Performance, and Home Prices on the tabs below. Note: The Mortgage Originations tab shows characteristics of new loans as of the time they are made, while the Mortgage Performance tab shows payment-related activity after origination.
Data sources: McDash Analytics and CoreLogic.

Last updated February 26, 2016

Home Purchase

Figure 1

Monthly Average (Mean) of Borrowers’ Credit Scores at the Time of Origination
Combined, and separately for Federal Housing Administration (FHA)-Insured/Veterans Administration (VA)-Guaranteed and Conventional Home Purchase Mortgages, on a Scale from 300 (high credit risk) to 850 (low credit risk)

At a Glance

For all home purchase mortgages, mean credit scores trended up from 2007 to 2013 and remain elevated, despite falling a bit since 2010 within the Conventional and FHA/VA categories separately. The separate downward trends in each of those categories since 2010 had little effect on the overall mean score due to a 2010–2014 shift away from FHA/VA toward Conventional loans (shown in Figure 2), which, on average, are used by higher-score borrowers. In 2015, the FHA/VA share of home purchase mortgages rebounded a bit, but so did the mean credit score in each of the two categories. As a result, the overall mean credit score for the District edged up from 739 in October 2014 to 742 in October 2015.

Figure 2

Monthly Share of New Home Purchase Mortgages Insured by the FHA or Guaranteed by the VA

At a Glance

As the volume of mortgage lending fell after 2007, FHA-insured and VA-guaranteed loans became a much larger share of the remaining home purchase mortgage market, accounting for the majority of home purchase loans in the McDash Analytics data by 2009. Their market share trended down from nearly 60 percent in 2010 to lows near 25 percent in late 2014 before rebounding slightly to 28 percent as of October 2015.

Home Refinance

Figure 3

Monthly Average (Mean) of Borrowers’ Credit Scores at the Time of Refinancing
Combined, and separately for FHA-Insured/VA-Guaranteed and Conventional Home Refinance Mortgages, on a Scale from 300 (high credit risk) to 850 (low credit risk)

At a Glance

Mean credit scores for refinancings rose significantly after 2007 to peak levels in 2010–2012. Mean scores then fell sharply, overall and within the Conventional and FHA/VA categories, during most of 2013, as refinancing volume fell somewhat more quickly among high-score borrowers than among low-score borrowers. Since late 2013, mean scores and volumes have continued to fluctuate together, but on a smaller scale. Volumes of and scores for mortgage refinancings rebounded slightly from mid-2014 into early 2015 and then slipped again. Scores remain well below peak but above pre-2008 levels.

Figure 4

Monthly Share of Mortgage Refinancings Insured by the FHA or Guaranteed by the VA

At a Glance

FHA/VA loans represented over one-third of all Ninth District home refinance mortgages at the peak of the 2008 financial crisis. Their market share dropped below 15 percent by mid-2011 and has fluctuated in a 6 to 12 percent range since late 2012. The share as of October 2015 was about 8 percent.

The Incidence of Mortgage Problems

These calculations are based on the monthly number of loans identified (by their mortgage servicer) as being in varying stages of delinquency, foreclosure, or post-foreclosure status. For each stage, the number of problem mortgages is then divided by the total number of active loans for each month to produce the respective rate. The data in this section are through December 2015.

Note: For the Montana and North Dakota counties shown in white on the maps below, results are not available due to insufficient data.

Figure 5

Early-Stage (30–89 Days) Delinquency Rate by Month

At a Glance

Early-stage delinquency rates in the Ninth District peaked in late 2008 and have generally trended down since then. These delinquency rates typically fall early in each year and then trend up through year-end. They did so again in 2015, falling to 2.0 percent in March but ending the year at 2.2 percent, only very slightly below a year earlier. Nonetheless, December’s rate was lower than in any December since 2005.

Figure 6

Current Early-Stage (30–89 Days) Delinquency Rate by County

At a Glance

By Ninth District standards as of December 2015, early-stage delinquency rates are moderately high in much of the eastern and far western portions of the Ninth District, including many rural and ex-urban counties. The highest and lowest rates in the Ninth District are concentrated in counties in the Dakotas and central to eastern Montana that mostly have relatively few active loans.

Figure 7

Late-Stage (90+ Days) Delinquency Rate by Month
(Excludes Foreclosed Mortgages)

At a Glance

Late-stage mortgage delinquencies fell noticeably in the first half of 2015 but then stabilized near December’s rate of 0.87 percent. As a result, the District’s average late-stage delinquency rates in the second half of 2015 were, on average, the lowest since 2007. Nationally, these delinquency rates were higher but followed a similar trajectory in 2015.

Figure 8

Current Late-Stage (90+ Days) Delinquency Rate by County
(Excludes Foreclosed Mortgages)

At a Glance

As of December 2015, late-stage delinquency rates tend to be relatively low in the north central section of the Ninth District (in much of North Dakota and nearby portions of South Dakota, Montana, and Minnesota). Elsewhere, counties with relatively high rates are mostly outside of the District’s largest metropolitan centers.

Figure 9

Foreclosure Rate by Month

At a Glance

Ninth District foreclosure rates continued to ease slightly in 2015, reaching 0.55 percent in December. Although still more than double the rate in 2004, it’s the lowest Ninth District rate since April 2007. The national foreclosure rate remains higher but fell somewhat faster in the 12 months ending in December 2015.

Figure 10

Current Foreclosure Rate by County

At a Glance

By Ninth District standards as of December 2015, mortgage foreclosure rates were moderately high to high in the Ninth District’s Wisconsin counties and in an array of mostly rural or exurban counties across the rest of the Ninth District. Many of the Ninth District’s counties with low foreclosure rates are in nonmetropolitan areas of North Dakota, South Dakota, and eastern Montana and often have relatively few active loans.

Figure 11

Rate of Real Estate Owned (Post-Foreclosure) by Banks, by Month

At a Glance

The real estate owned (REO) rate in the Ninth District drifted downward in 2015, reaching 0.75 percent in December, its lowest level since April 2007. The national rate has changed little, on net, since late 2011, and the Ninth District rate has fallen to parity with it over that period. (Some sharp changes in the rate, such as in 2009 and 2011, may be due in part to changes in the composition of the group of servicers reporting data.)

Figure 12

Current Rate of REO by County

At a Glance

As of December 2015, Ninth District counties with relatively high REO rates lie mostly in rural portions of the eastern half of the Ninth District.

Figure 13

Home Price Index (Year 2000=100)

NOTE: DATA IN THIS SECTION ARE THROUGH NOVEMBER 2016.

At a Glance

The price index reported is for the entirety of each state (rather than the Ninth District portion, in the cases of Michigan and Wisconsin). The dashed line shows the national price index.

Data Overview

Figures on the Mortgage Origination and Mortgage Performance tabs reflect Federal Reserve Bank of Minneapolis staff calculations based on data on first-lien mortgages provided by McDash Analytics (formerly Black Knight Financial Services). These data are derived from the servicing portfolios of residential mortgage servicers, including many of the largest servicers in the U.S. The figures shown are based on the unadjusted loan data in the McDash Analytics source databases, which account for a substantial share of all mortgages but cannot be guaranteed to be representative of all mortgages in each geography. As of January 2014, McDash Analytics provided monthly loan-level information on approximately 31 million active loans in the U.S. McDash Analytics estimated that, as of December 2012, its dataset covered 71 percent of the first-lien mortgage market. However, coverage can change over time and varies by date, type of mortgage, and geography.

Figures on the Home Prices tab reflect Federal Reserve Bank of Minneapolis staff calculations based on Home Price Index (HPI) data provided by CoreLogic. The HPI measures the change in home prices using a repeat sales methodology.


Have questions or comments about this page?
Please e-mail them to mpls.communitydevelopment@mpls.frb.org.