This article appeared in the March 2010 issue of Current Economics with permission of the author.
After nearly three years of falling home sales, plummeting house prices and moribund construction activity, 2009 saw signs of stabilisation and recovery in the US housing market. Key housing market variables have been so volatile in recent years that it is easy to lose track of what a recovered housing market would look like. Home prices rose by 42% between 2003 and 2006, then fell by 30% between 2006 and 2009. Housing starts peaked above 2.0 million in 2005 then fell to 550 thousand in 2009. Not one of these numbers is anywhere near long-run sustainable levels.
At their current value, home prices are more likely to be modestly undervalued than significantly overvalued and at just over 500,000, housing starts are running not only below the long-run pace of household growth, but barely enough to keep pace with depreciation.
The outlook for the housing market in 2010 will be struggle between forces pulling the housing market back towards long run levels and those hindering this movement.
Supporting growth will be renewed job creation and the improvement in housing affordability, while obstructing growth will be the unwinding of temporary stimulus and the looming supply of foreclosures. Under the weight of these countervailing forces, the housing market is likely to move sideways in 2010, before beginning a multi-year recovery in 2011. In a recovered housing market, home price growth should return to roughly 5% annually – a pace consistent with growth in income and the cost of new construction. New housing construction should rise to 1.5 million, in line with the growth in households.
The homebuyer tax credit contributed to the rebound in home sales over the course of 2009, but after a last minute extension by Congress in November, the credit is set to expire in April of this year. Baring another extension, the expiration of the tax credit will become a drag on home sales in the remainder of 2010. How much of a drag in part depends on how much of a boost the tax credit provided. Because the tax-credit is temporary, it draws forward sales that would otherwise have taken place later in the year (or subsequent years). Since the original tax credit applied only to first-time homebuyers (buyers who hadn’t owned a home in the past three years), one approach to estimating the impact is to look at the change in the share of new homebuyers in total home purchases. First-time homebuyers typically represent 40% of the housing market according to the National Association of Realtors (NAR), but represented 45% of the existing homes purchased in 2009. A five percentage point swing in the share of new homebuyers, if fully attributable to the first-time homebuyer tax credit, implies an additional 230,000 home sales.
Since the tax credit acts like a price cut, another way to quantify its impact is to consider how housing demand responds to a change in price (holding all else constant). An $8,000 tax credit is equivalent to close to a 5% reduction in the median priced home. Economic studies generally report that a 1% change in home prices results in less than a 1% change in housing demand1. However, given higher down payment requirements, the tax credit also acts to alleviate an important credit constraint on homebuyers. Loan-to-value ratios have fallen dramatically over the course of the housing bust due to more cautious lending behavior. An added 5% to 10% down payment, in this environment, is likely to cause a larger demand response than a simple 5% reduction in price.
Given the direct price impact and the additional liquidity driven boost, the tax credit likely led to additional 250,000- 300,000 home sales over the course of 2009. The extension of the tax credit until April (and its expansion to include existing home owners), could therefore lead to an additional 30,000 to 60,000 home sales in the remaining months before April. Once the tax credit expires, the additional home sales that it stimulated will drop off. If December and January’s sales declines are any indication, this could happen fairly abruptly in the months that follow.
The unprecedented rise in the rate of home foreclosures has made the current housing bust unique. As of the fourth quarter of 2009, more than 4.6 million mortgages were at least 30 days behind in their payments and more than two million were in foreclosure. Foreclosures are particularly acute in the regions of the country that saw the worst lending excesses over the course of the boom. The states of California, Florida, Arizona, and Nevada alone represent 40% of seriously delinquent loans (loans in foreclosure or more than 90 days delinquent). The Obama administration has recently announced a $1.5 billion program to help housing finance companies in these states hit hardest by the crisis. Delinquencies and foreclosures are also highest among the riskiest mortgage types. As of the fourth quarter of 2009, 42.7% of all subprime adjustable rate mortgages (ARMs) were seriously delinquent, compared to 5.0% of prime fixed rate mortgages.
Fortunately, there have been signs of stabilisation in both the most severely hit regions, as well as the worst performing mortgage types. Foreclosure filings, according to RealtyTrac, fell by 10% in January and fell by even greater percentages in both California and Florida2. In California, months supply of unsold homes fell dramatically over the course of 2009 to under 4.0 months by December, and median home prices ended the year 8.4% higher than a year earlier.
In terms of loan performance, the rate at which subprime loans are becoming seriously delinquent has also slowed over the last several quarters. Stabilisation in the lower quality loans is a good sign since these are the most likely to have been purchased at the peak of the housing boom and are therefore the deepest underwater.
The main force behind a household falling behind in their mortgage payments is conditions in the job market. The 30-day delinquency rate has fallen in each of the last two quarters, reflecting a much slower pace of job losses than earlier in the year. As job creation picks up in the next few months, the number of mortgages entering the initial stages of delinquency will decline further. Nonetheless, the heightened supply of delinquent loans implies that there will be more foreclosures in the pipeline in 2010. In several states, moratoria on foreclosure proceedings has led to an increased pipeline of loans delinquent 90 days or more, which could well become foreclosures in the next year. In fact, the real problem in terms of foreclosures is the deterioration in mortgages once they have entered delinquency. Prior to 2006, for every one mortgage 60-days delinquent there were roughly 4 mortgages 30-days delinquent. By 2009, this ratio had risen to 1 in 2. The rising ratio implies a higher proportion of loans entering delinquency are remaining in delinquency (and eventually foreclosing). At 49% of the total, mortgages that are at least 90 days past due are now leading the pack.
One of the brakes on foreclosure growth through 2009 was the Obama administration’s $75 billion Home Affordable Modification Program (HAMP). By the end of 2009 HAMP had led to trial loan modifications on close to 1 million mortgages. The program has two main elements. The first is to offer refinancing on mortgages owned by government-sponsored enterprises, Fannie-Mae and Freddie-Mac, where the loan-to-value ratio has risen above 80%. The second part of the program is to offer incentives to lenders to lower interest rates on mortgages where monthly payments exceed 40% of monthly income. While the initial uptake of the program is promising, so far only a small portion of these modifications have been made permanent. However, this too is improving, and in January the number of permanent loan modifications nearly doubled to over 116,000, or roughly 12% of total loan modifications.
The Treasury has also initiated a program to incentivize borrowers and loan servicers to engage in voluntary “short sale” or “deed-in-lieu” agreements when a loan modification is not possible. In a short sale, the borrower voluntarily sells the home and gives the proceeds to the lender. Even without the Treasury program, short sales are in many cases the best option for both parties involved. Most foreclosed homes are eventually owned by the lender and often lie vacant until a buyer can be found. In these conditions, homes deteriorate and sell at a larger discount. For borrowers, a short sale avoids the pain and credit woes of a full foreclosure. While it remains uncertain how effective the plan will be in stemming the tide of foreclosures, a number of mortgage lenders have announced plans to offer short sale options to delinquent borrowers in recent months.
While the sheer size of the pool of delinquent loans can seem overwhelming, it is important to note that a portion of foreclosures and short sales can be absorbed by the market. In fact, according to NAR, of the 5.1 million home sales in 2009, more than 20% were foreclosed sales, and more than 12% were short sales. The improvement that took place in home prices in 2009 showed that even with distressed sales making up more than 30% of sales (roughly 1.5 million), home prices could still stabilize.
Currently, there are 4.6 million homes in delinquency, of which close to 2 million are 90 days or more past due. Given that there are currently 3.6 million homes for sale, the addition of 4.6 million homes on the market would more than double the inventory of unsold homes. Even allowing for the pace of sales to rise in 2010 to its pace in the late 1990s, such an increase would push the month’s supply of unsold homes well past its peak in late 2008. This would, in all likelihood, lead to a further significant decline in home prices from their current level. In a best case scenario, foreclosure mitigation efforts and an improved job market remove the majority of the currently delinquent loans from the pipeline and home prices continue on an upward trajectory driven by rising demand.
Perhaps the most likely scenario is in between these two extremes. Should the pipeline of potential foreclosures be cut in half, we would see only a modest increase in months supply. While some month-to-month volatility should be expected, prices in this situation would more or less stabilize around their current level.
The current housing market downturn is without a doubt the longest and deepest in modern times. In amongst the carnage it is easy to lose sight of the fact that Americans still need houses to live in and that the number of Americans grows by close to 3 million every year. Of course, what matters for the housing market is the growth in the number of households; which, in addition to population growth, is determined by conditions in the economy. According to the US Census Bureau, there were 398,000 new households formed between March 2008 and March 2009, down from 772,000 over the same period in 2008 and 1.6 million in 2007. The fall in the growth of households in 2008 and 2009 was due to both an outright decline in the number of households headed by younger aged individuals as well as to a slowdown in the growth rate of older cohorts (even while the population of both these groups continued to grow). As shown in the chart, the propensity to form a household (often referred to as the “headship rate” and calculated by dividing the number of households in each cohort by its population) rises with age, with the biggest increase occurring in the 25- 29 age cohort. Even smoothing out some of the year-to-year volatility, the headship rate of younger cohorts has fallen over the last few years, and this has shown up in terms of a slower pace of household formations.
The decline in the headship rate, especially of younger cohorts reflects the deterioration in income and job prospects that has taken place over the course of the recession. As we argue in our report, “US Won’t Have a Jobless Recovery,” positive job growth is just around the corner. We expect the US economy to add 2.2 million jobs over the course of 2010. While the unemployment rate will remain relatively elevated, it fell below 10.0% in January and remained there in February. Even with movement of discouraged workers back into the labor force, the unemployment rate will likely move further downward as the year progresses.
The swing from negative to positive job and income growth will go a long way in pulling the rate of household formations towards its long run growth rate of 1.4 million. We do not expect new home sales to rise to 1.4 million in 2010, but given the current level of housing construction, sales need only rise a portion of this amount in order to move the month’s supply of new homes to historical levels. Home builders have cut back housing construction to the point that even at the current rate of sales, inventories of unsold new homes are declining. Currently there are only 230,000 new single-family homes for sale – the lowest inventory in over 40 years. Despite the setback in January, given our outlook for jobs, new single-family home sales are likely to rise to half a million by the end of 2010.
The majority of homes are bought and sold on the existing housing market. The turnover of existing homes – as measured by the ratio of existing home sales to the total stock of homes – rose considerably over the course of the housing boom, reaching 7.4 home sales for every 100 homes in existence in 2005, before falling to a low around 5.0 in 2008 and 2009. As the job market improves through 2010, it will allow more homes to change hands again. Nonetheless, it should be remembered that the heightened turnover of the boom years was also due to speculative buying and selling and sales are unlikely to reach these highs again. In all likelihood, the rate of housing turnover will return to its late 1990s level when market conditions were relatively balanced. Taken together with our forecast for employment and new housing construction, this implies existing home sales of just under six million in 2010.
Another factor in returning housing demand to more normal levels is the relationship between home prices and household income and between home prices and rents. In the long-run, the value of a home should reflect the present value of the future stream of housing services it provides. The higher are household incomes, the more housing services households will demand. In relatively competitive market, this will also equal the value of rents. By both of these metrics, home prices were dramatically over-valued over the course of the housing boom, but with the decline in prices since 2006, prices have since returned to at least historical fair market value. This gives us confidence that while there may be some volatility in home prices due to the unwinding of stimulus and the supply of foreclosures on the market, as long as the broader economic recovery continues, home prices are unlikely to see further significant declines going forward.
The other factor for housing affordability is the outlook for mortgage rates. After holding relatively steady through most of 2008, 30-year fixed rate mortgages finished 2009 at 4.9%, a drop of nearly a full percentage point. Federal Reserve purchases of mortgage backed securities and agency debt were an important element of this decline.
By early 2010, the interest rate spread between 30-year mortgages and 10-year government bonds had fallen by more than 150 basis points (1.5 percentage points). The Federal Reserve has announced that it will cease its mortgage purchases in the second quarter of this year; however, it is unlikely to begin selling these assets in the near future. Given our expectation for government bond yields and a movement in mortgages spreads towards more normal levels, mortgage rates will move upward over the course of 2010, likely giving back the percentage point decrease that took place over 2009.
Taken together with a stabilisation in home prices, further improvements in affordability will depend on rising income growth.
Existing Home Turnover*
While it is difficult to determine how many distressed sales will enter
the market in 2010, we can say with a fair degree of certainty that the
number of homes entering delinquency is likely to fall in 2010 as the job
market improves. Loan modifications will also help to stem the tide of foreclosures
as will creditors’ decisions to stay out of the market until the home
price outlook improves. The unwinding of tax stimulus and rising mortgage
rates will take some wind out of home sales through the second half of the
year, but by then the main force behind the housing recovery will be renewed
job creation and income growth. Importantly, the 30% decline in home prices
has eliminated the gap that rose during the housing boom between the price
of homes and its underlying determinants of income and rents. Moreover,
the drastic cuts to housing supply mean that it is only a matter of time
before underlying demographic driven demand begins to exert upward pressure
All told, while headwinds are likely to slow the housing recovery through 2010, they are unlikely to be strong enough to derail it altogether. By early 2011 the stage should be set for a fairly vigorous housing recovery. As a result of the low base from which it is starting, residential construction is likely to rise by close 20% from 2010 levels, adding close to a percentage point to total GDP growth.
1 See Kahn, James A. "Productivity Swings and Housing Prices". Current Issues in Economics and Finance, Federal Reserve Bank of New York. Volume 15, Number 5, July 2009.
Krainer J. "Housing Markets and Demographics". FRBSF Economic Letter, Number 2005-21, August 26, 2005.
2 Tanzi, Alex. "US Home Foreclosures Rose 15.1 Percent in January," Bloomberg LP. February 11, 2010.
To download free samples of our publications, featuring forecasts and analysis covering over 80 countries, please click below:
For further articles from Current Economics, please see News and Articles.