Recent housing data have been generally been encouraging. However, the large number of residential properties that are "underwater"—meaning the borrower owes more on the mortgage than the property is worth—casts a long shadow on the sustainability of the housing recovery. Goldman estimates that approximately 10 million properties are currently underwater. Although this number has not changed much during the past three years, there is much divergence across the nation: California, Michigan, and Arizona, for example, experienced significant improvement, while Georgia, Utah, and Missouri saw many more properties falling underwater during this period.
[6]
Even though the number of underwater properties stayed roughly unchanged at the national level over the past three years, different states experienced very different developments over this period of time. Exhibit 2 ranks states by the change in the percent of first-lien mortgages with negative equity from April 2009 to April 2012. Among the 31 states that had at least 0.5 million outstanding first-lien mortgages in April 2009, California, Michigan, and Arizona saw the largest improvement. For example, 42% of residential properties with mortgages in California were underwater in 2009, and that number came down to 29% in 2012. This trend likely resulted from both stabilizing house prices and the relatively fast speed at which foreclosed underwater properties are cleared out of the system. On the other end of the spectrum, Georgia, Utah, and Missouri experienced the largest deterioration in their negative equity problem. For example, 24% of residential properties with mortgages in Georgia were underwater in 2009, but that number increased to 42% in April 2012. This trend is largely caused by house prices falling more in these states than other states during the past three years.
[7]
What are the potential impacts of negative equity on the macro economy? Broadly speaking, negative equity affects the economy in two ways.
First, negative equity weighs on consumption. As shown in Dynan (2012, see full reference below), highly leveraged homeowners had larger declines in spending between 2007 and 2009 relative to other homeowners, and leverage appeared to weigh on consumption above and beyond what would have been predicted by housing wealth effects alone. The large number of underwater properties that we have estimated above suggests that many homeowners are still highly leveraged today. As a result, the deleveraging process they are going through is likely to drag down consumption and aggregate demand during the recovery.
Second, borrowers with negative equity are more likely to default on their mortgages. Both theoretical research, such as Campbell and Cocco (2011), and empirical research, such as Bhutta, Dokko, and Shan (2010), show that mortgage default rates increase significantly when borrowers become deeply underwater.
Given that there are 3 million first-lien mortgages that have LTVs of 125% or above as of April 2012, whether or not a large fraction of these mortgages will default in the near future has important implications for the housing market recovery.