Thursday, January 22, 2015

What Effect Will Lower FHA Mortgage Insurance Premiums Have on First-Time Homebuyers?

by Dan McCue
Research Manager
As mentioned in Tuesday’s State of the Union address, the Federal Housing Administration (FHA) announced that it will lower costs of government mortgages by reducing the annual Mortgage Insurance Premium (MIP) rate on most of its new single family home loans by 50 basis points, beginning on January 26. This left many wondering what exactly this means for borrowers and, in particular, what effect it might have on first-time homebuyers who make up a high portion of FHA borrowers and who largely remain on the sidelines.   

The MIP is a fee that FHA charges borrowers to maintain a reserve of funds necessary to insure lenders against losses on its loans. Since the MIP cost is assessed annually as a percent of outstanding principal and then divided across 12 monthly payments, the effect of the drop in MIP rate is in many ways similar to a change in the interest rate of a loan. But the MIP payment is added on top of monthly principal and interest payments, so it does not factor into amortization schedules in quite the same way as interest rates. Additionally, unlike interest payments that change month to month as a borrower pays down the loan balance, the MIP amount is also recalculated just once a year and remains fixed for that time. FHA also charges an up-front mortgage insurance premium (UFMIP) that is a one-time fee incurred at the origination of the loan, which remains unchanged by the recent announcement and currently stands at 1.75 percent of the original loan balance for most FHA loans.

This recent move to lower the MIP comes after several rounds of increases made to cover losses to FHA loans following the housing bust. Even after tightening lending standards and four rounds of MIP increases starting in 2010 (see Table 1), FHA was still forced to draw $1.7 billion from the Treasury in 2013 to remain solvent.  However, FHA’s FY2014 annual report, released in November, shows the FHA Mutual Mortgage Insurance Fund grew by $6.1 billion last year to a positive $4.8 billion in value. Although the capital ratio remains at just 0.41 percent, still well below the legally-mandated capital ratio of 2 percent, the tide has apparently turned and, as pointed out in a recent Urban Institute analysis, financial projections must suggest there is room for FHA to reduce MIP on new loans while still growing its reserves. Indeed, as we see in table 1, the new, lower MIP rates are still well above levels prior to October 2010.

Table 1: History of FHA mortgage insurance premium changes

Date of Change
<=95.0% LTV
> 95.0% LTV
FHA Announcement
Prior to October 4, 2010
50 bps
55 bps
October 4, 2010
85 bps
90 bps
Mortgagee Letter 10-28
April 18, 2011
110 bps
115 BPS
Mortgagee Letter 11-10
April 9, 2012
120 bps
125 bps
Mortgagee Letter 2012-4
April 1, 2013
130 bps
135 bps
Mortgagee Letter 2013-4
January 26, 2015
80 bps
85 bps
Mortgagee Letter 2015-01

Note: Rates shown are for 30-year mortgage that is no greater than $625,500.

So what is the impact of this change? The 50 basis point reduction in the FHA MIP rate amounts to a difference of $500 per year, or about $42 a month for every $100,000 of mortgage balance. Based on the current median home sales price, the administration announced that the average borrower will see a reduction in costs of about $900 per year. Likely this will have some impact at the margins. For first-time homebuyers, to whom fully 75 percent of FHA loans in FY2014 were originated, the modest increase in affordability could result in a short-term bump in sales activity that may or may not be measurable. 

To see how this change may affect first-time homebuyers, we look at the impact on home purchasing power for the typical renter. According to data from the Census bureau, the median renter household had an income of about $34,000 at last measure in 2013. Assuming this renter obtains an FHA mortgage with 3.5 percent down, a 4 percent interest rate and maintaining a 31 percent front end debt to income ratio – which is the published limit for FHA manually underwritten loans - the change in MIP increases the amount of house they could potentially afford from about $152,000 to $163,000, or about 7 percent. This price is still well below the most recent published NAR median home sales price of $205,300, but nonetheless amounts to an increase of $11,000 in home purchasing power for this borrower – also equivalent to a 6.7 percent drop in home price – that may open up a few more buying opportunities depending on the number of homes available on the market in this price range. The administration announcement estimated this at approximately 250,000 additional buyers over three years. 

What is not addressed by the analysis of the new MIP rates is the extent to which they will help those with less than stellar credit. FHA lenders use credit overlays to narrow the field of potential borrowers to those with the highest credit ratings. These and other credit barriers – evidenced by the still-historically elevated median credit score of the typical FHA borrower - are likely much more limiting to first-time homebuyers than affordability of mortgage insurance payments.  At the same time, rising home prices and/or interest rates may prove to have much more of an impact on affordability over the coming months. At the end of the day, however, the result is that this is a modest step that most likely won’t jeopardize the financial standing of FHA but will increase mortgage affordability for thousands of FHA borrowers and potentially even increase home buying opportunities at the margins at a time when credit remains tight, home sales are sluggish, and first-time home buying is struggling - which seems like a step in the right direction.      

Thursday, January 15, 2015

Growth Slowing in Home Remodeling in 2015 (+ New Remodeling Report January 29)

by Abbe Will
Research Analyst
As the broader housing market continues its sluggish recovery, growth in home improvement spending is also expected to soften throughout the coming year, according to the Joint Center's most recent Leading Indicator of Remodeling Activity (LIRA) released today. The LIRA projects annual growth in home improvement spending will decelerate from 6.3% in the first quarter of 2015 to 1.6% by the third quarter.

Due in part to weakening home sales last year, growth in remodeling spending is expected to deflate somewhat in 2015. Homeownership rates continue to slide as lending remains tight and first-time homebuyers are not yet returning to the market.  

Although contractor sentiment has cooled in recent quarters, it remains favorable overall. House price gains are moderating but still strong and home sales appear to be turning a corner now, all of which bodes well for continued, if more moderate, home improvement gains for 2015.

NOTE ON LIRA MODEL:  Beginning with the first quarter 2014 release, long-term interest rates were removed from the LIRA estimation model.  For more information on the reasons for and implications of this change, please read our blog post from April






 NEW REMODELING REPORT 
& LIVE WEBCAST – JANUARY 29

On Thursday, January 29, the Joint Center for Housing Studies will release its latest biennial report on the remodeling industry, Improving America’s Housing: Emerging Trends in the Remodeling Market.  Almost fully recovered from the recent downturn, the report identifies the remodeling industry segments that will support further growth in the years ahead. Please join us for the live webcast release at 12:00 p.m. Eastern.  

Tuesday, January 13, 2015

What Loss Mitigation Taught Us About Housing Finance Reform

by Patricia McCoy
Guest Blogger
From time to time, Housing Perspectives features posts by guest bloggers. Today's post was written by Patricia McCoyLiberty Mutual Insurance Professor, Boston College Law School, and former Assistant Director for Mortgage Markets, Consumer Financial Protection Bureau.

Since 2007, the federal government and servicers have groped toward striking the right balance between cost-effective loss mitigation and unavoidable foreclosures for homeowners with delinquent home mortgages. Among other things, this painful experience resulted in a cornucopia of data about the right way and wrong way to do loss mitigation. Nevertheless, none of the leading housing finance reform proposals has incorporated these lessons. Take, for example, the Johnson-Crapo bill, which was the leading reform contender and made it to the Senate floor. That bill would vaguely require servicers to establish “loss mitigation options that seek to enhance value” but says nothing about the best way to do so. That oversight is unfortunate because it sets us up to repeat the mistakes of the past.

First, some history. Eight years ago, the federal government became focused on foreclosure prevention as mortgage delinquency rates began to spike. The George W. Bush Administration sought to achieve that goal through moral suasion and voluntary compliance by servicers under the aegis of HOPE NOW. Later, the Obama Administration turned up the heat by offering a carrot and a stick: a carrot consisting of the HAMP Program, which paid servicers to grant loan modifications when doing so would increase recovery, and a stick through rulemakings and enforcement. One of the results was a rich trove of data on what makes loss mitigation work and why.

So what did we learn? I address this question in some detail in a chapter in the Joint Center’s latest book, Homeownership Built to Last, but here are a few quick takeaways:
  1. Standardized decision tree, or “waterfall,” such as that employed by HAMP to lower monthly payments, is key to minimizing default rates (by cutting the interest rate or, if necessary, reducing principal).

  2. Foreclosure prevention is more successful the sooner it is granted after a homeowner’s first delinquency – ideally within two to three months.

  3. The federal government should offer meaningful relief to people who are temporarily unemployed and need help making their mortgage payments until they get back on their feet.
These three techniques are a win-win for distressed homeowners and for investors by avoiding needless foreclosures while maximizing investor recovery.

But there’s one final lesson that is directly relevant to housing finance reform. We can’t implement lessons one through three unless we remove servicer barriers to effective foreclosure prevention. In 2007 and 2008, the private-label mortgage-backed securities market collapsed and that market remains moribund today. In the aftermath, servicers rushed to foreclosure in too many cases, saying that pooling and servicing agreements with investors (PSAs) tied their hands. Investors disputed those claims, complaining that the PSAs did not in fact bind servicers’ hands and that foreclosure prevention would have enhanced recovery in many of those cases. 

After that sour experience, investors will not be eager to rush back to the private-label market unless loss mitigation rules give them stronger protections. It is important, going forward, that PSAs in future deals give servicers no excuse to deny loan modifications that will increase investor recovery vis-à-vis foreclosure. Since PSAs are privately negotiated, the only real way to assure that is to prescribe standard loss mitigation protocols and require PSAs to follow those protocols in any housing finance reform law that Congress enacts. The protocols should require servicers to evaluate loan modification requests by distressed borrowers using a standardized waterfall that is designed to reduce monthly payments to an affordable target level. Under that waterfall, servicers should attempt to attain the target payment level first through interest rate reductions and then, if need be, through principal reductions. The statutory protocols should also set time limits for loan modification decisions to help encourage early intervention. With these protocols in place, we can help avoid the experience in recent years where useless foreclosures pushed down home prices and delayed the economic recovery.

Monday, January 5, 2015

11+ Million Undocumented Immigrants in the U.S. Could Be Important for the Housing Recovery

by George Masnick
Senior Research Fellow
President Obama’s recent announcement that he will take executive action on immigration could be an important step in further supporting the sluggish housing recovery. Immigrants have historically been an important part of housing markets, especially for the past two decades. The foreign born have accounted for about one third of net new household formations since 1994 (Figure 1). Immigrants owned 11.2 percent of all owner occupied units in 2014, up from 6.8 percent in 1994, accounting for 27.5 percent of owner household growth over this period.


Source: Joint Center Tabulations of 1994 and 2014 Current Population Survey Data

For households under age 45, the foreign born have accounted for virtually all household growth over this period, as the aging of Generation X led to a decline in native born households in this age range. Meanwhile, as immigrants accounted for a similar level of household growth among heads over age 45, the overwhelming majority (80.3 percent) of total growth in this age range over the past 20 years is attributable to native-born adults. The particularly high numerical household growth among 45-64 year olds represents the aging of the baby boom generation into this age span during this 20 year period. 

Households headed by those under the age of 45, however, have had a very different mix of growth by nativity of the head. Figure 2 breaks household growth for under-45 year olds into 5-year age groups, and separates the native-born growth into natives with one or both parents being foreign born (second generation immigrants) and natives where both parents were native born. I have separated out the native-born children of immigrants to make the point that immigrant population growth has both an immediate impact on housing markets and a secondary impact (approximately 20-40 years later) through the children they bear in the U.S. As shown, second generation immigrants have accounted for the largest share of growth in under-30 year old households, followed by immigrants themselves. Looking forward, we expect that second generation growth originating from the higher immigration levels of the 1990s and early 2000s will be even larger.

Source: Joint Center Tabulations of 1994 and 2014 Current Population Survey Data

Over the next two decades, native-born cohorts that are third or higher generation immigrants will begin to exercise a more positive influence on household growth as millennials born after 1985 continue to age into young adulthood. And as the smaller Generation X enters middle age, immigrants and their native-born children will continue to bolster households in this age range, just as they did over the past two decades for the under-45 age group.

There is considerable potential for additional household formation and homeownership among the foreign born, particularly for the nation’s estimated 11 million undocumented immigrants, many of whom have been cautious about signing a lease or applying for a mortgage given their status. A recent Pew Research Center report estimates that 61 percent of undocumented immigrants have been here for 10 or more years, and the longer the duration of residence in the U.S., the more likely that immigrants will form independent households and pursue homeownership.  

Because the data in the above charts are national in scope, and because undocumented immigrants are more concentrated in some parts of the country than in others, their impact at a local level can be much more pronounced. While there are obviously a number of factors to weigh in deciding whether and how to offer a path to citizenship for undocumented immigrants, a greater appreciation of the role that immigrants play in our housing markets—and  through housing in the health of the overall national economy—should be a part of the discussion of immigration reform. Granting undocumented immigrants who have been working here for an extended period of time, especially those with children born in the U.S., the opportunity to remain in this country under the law will certainly be an important boost to the housing sector of the economy. 

Wednesday, December 17, 2014

3 Facts about Marriage and Homeownership

by Rachel Bogardus Drew
Post-Doctoral Fellow
In conversations about the declining homeownership rate in the U.S., some commentators have pointed to declines in the share of married people as an important contributing factor. To be sure, married couples are much more likely to be homeowners than unmarried individuals, due to their generally better socio-economic status. Married couples are also more likely to have children, and therefore more likely to want larger homes in areas with more family-friendly amenities such as safe neighborhoods and good schools. And these features are more often found in suburban neighborhoods where a larger percentage of the housing stock is available to own rather than rent. A decrease in married couples, goes the argument, should reduce the share of households that own their homes. While this is true, all else equal, this rationale fails to consider three important facts about marriage and homeownership that suggest this effect is not as large as many people imagine, and that recent trends are not necessarily indicative of a fundamental shift in demand for owned homes.

Fact #1: Marriage rates have been declining for decades.

While concern over the connection between declining marriage rates and declining homeownership has gained traction in recent years, the share of households headed by a married couple has actually been on a steady downward trend since the 1960s when over 70 percent of adults ages 18 and older were married (Figure 1). Declines since 1960 in the married share have been dramatic, falling fully 17 percentage points from a half-century earlier. The time frame of this decline, however, coincides with a period of steady increases in the national homeownership rate, from 62 percent in 1960 to 66 percent in 2000. Indeed, only during a brief period in the early 1980s did the homeownership rate decline slightly, at the same time that marriage rates were falling (though moderating somewhat from more dramatic decreases during the 1970s).

Since the turn of the century, of course, the homeownership rate has declined from a peak of 68 percent in 2004 to 64 percent earlier this year. In that time, the married share of the population has also fallen slightly, thus fueling the argument that the two trends are linked. Yet these recent declines in homeownership and marriage rates have also been greatly impacted by conditions in the macro economy, particularly the slow pace of earnings growth and high unemployment. The decline in marriage rates is also clearly a continuation of a long-term trend related to socio-demographic changes in the population and shifting social norms around relationships and cohabitation. The fall-off in homeownership, meanwhile, is largely viewed as a reversion to conditions that existed prior to the late 1990s and early 2000s spike in home buying, rather than purely a reflection of changing household compositions. This is not to say that there is no connection between marriage and homeownership, only that there is more to this story. As the statisticians say, correlation is not causation, so looking at recent changes in marriage and homeownership does not necessarily mean that one fully explains the other.


Note: The homeownership rate from 2000-2014 is calculated by using the Decennial Census rate from 2000 and adjusting annually by the change in the homeownership rate reported by the Housing Vacancy Survey.
Sources: U.S. Census Bureau’s Decennial Census and Housing Vacancy Surveys.

Fact #2: People are still marrying, just later.

The share of married adults in the U.S. only tells part of the story about the marital status of Americans. A better indicator may be the share that have ever been married, observed in mid-life when most people are settled and those who will marry have likely done so. By their late 40s, more than 85 percent of Americans report having been ever married, with that share leveling off over the last decade (Figure 2). Even by their late 30s, nearly 80 percent of people have tied the knot. The big change over the last 30 years, however, has been in the share of young adults aged 25-34 who have not married, which rose from 25 percent in the mid-1980s to almost 50 percent today, with notable acceleration over the last 10 years. The driving force behind this trend has been a steady increase in the age at first marriage, from 20 years old for women and 22.5 for men in the late 1950s, to 26.5 and 29 years old, respectively, in 2011. Young adults are more likely today to pursue post-secondary education, to relocate to a new area for employment, and to live with partners before marrying, all of which combines to delay the trip down the aisle. Again, these trends are not new, but have been gaining momentum for the past half-century.


Source: U.S. Census Bureau’s Decennial Census and Current Population Surveys.

It is worth noting that most young adults today still expect to marry eventually. The Pew Research Center reports that 66 percent of never-married adults under age 30 say they would like to get married someday. And even higher shares of young adults say they want to own a home; recent research published by the Joint Center shows that 88 percent of renters ages 25-34 plan to buy a home, and that their intentions are not influenced by their current marital status. It appears that, like marriage, many of these young adults are not rejecting homeownership, but rather delaying their purchase until they are more personally and financially settled.

Fact #3: It’s not only the current marriages that drive homeownership.

This last point about marriage and homeownership is one that does not get a lot of attention, but is very important to conversations about homeownership rates going forward. Among the unmarried population in the U.S. are a large number of previously married adults – those who are divorced or whose spouse has died – who have very different homeownership experiences from those who have never been married. Indeed, even as married households have been on the decline in the U.S., the share of householders who report being previously-but-not-currently married has remained steady at around 30 percent (Figure 3). The effect of these prior marriages on homeownership is profound, as previously-married householders are much more likely to own than never-married householders. Many of these are ‘legacy’ homeowners who bought while still married and remained in the home after becoming single, although some likely bought homes without a spouse, using proceeds from a home they owned when married. Previously-married householders have also seen a smaller decline in their homeownership rate, relative to currently-married householders, since the end of the housing boom a decade ago. All of these groups, moreover, have homeownership rates today that are at or above their pre-boom levels from the mid-1990s, suggesting that homeownership has been an important and popular choice for all households regardless of their marital status.


Note: Data are share of householders, not persons, by marital and tenure status. Households who report being separated from their spouse are not considered to be currently married.
Source: U.S. Census Bureau’s Current Population Surveys.

These facts make clear that recent concerns about the decline in marriage leading to less demand for homeownership are perhaps overstated. While married couples continue to own homes at higher rates than unmarried individuals, there is more to this relationship. Only recently have shares of married couples and shares of home-owning households both been declining, due mostly to economic factors, while vast majorities of the population still aspire to both marriage and homeownership. Nor are married couples the only ones owning homes, as both previously-married and never-married householders significantly increased their homeownership rates during the housing boom, with only small declines since the peak in 2005.

Finally, it is worth noting that most discussions of the role of marriage in housing decisions fail to consider the importance of other changes in household demographics that have had an impact on homeownership rate; the increasing share of minorities in the population, for example, can have a lowering effect on homeownership rates, while higher rates of college educated adults in the population should increase the share that own. The most important factors in recent declines in homeownership rates, however, are the performances of the housing market and economy, which determine whether households of all types are able to purchase homes. Stagnant incomes and constrained credit have had greater impacts on homeownership rates since the end of the housing boom than long-term demographic changes, and will likely continue to drive homeownership trends in the near future.