by Calculated Risk on 9/10/2009 02:50:00 PM

Professors Atif Mian and Amir Sufi (both University of Chicago Booth School of Business and NBER) published a new paper: The Household Leverage-Driven Recession of 2007 to 2009 This is related to their paper earlier this year: House Prices, Home Equity-Based Borrowing, and the U.S. Household Leverage Crisis, See: MEW, Consumption and Personal Saving Rate

A cross-sectional analysis of U.S. counties shows that areas with modest increases in leverage from 2002 to 2006 have experienced only a minor economic downturn, whereas counties with large increases in household leverage from 2002 to 2006 have experienced a severe recession. Our findings suggest that the process of household de-leveraging is likely to be the major headwind facing the economy going forward.

The authors have written a brief discussion of the paper at NPR Money: Lessons From The Fall: Household Debt Got Us Into This Mess

The Economist has a recent review of the paper: Withdrawal symptoms

More than a third of new defaults in 2006-08 were because of home-equity-based borrowing. Default rates for low credit-quality homeowners rose by more than 12 percentage points in places where housing was scarcest and prices had risen most. In “elastic” cities, by contrast, the increase was less than four percentage points. This suggests huge over-borrowing. Prospects for a sustained recovery look dim if households that are most inclined to spend are mired in negative equity.

Here are three graphs from the paper (posted with permission from the author):

Click on graph for larger image.

The first graph is figure 5B from the paper. Looking at the left panel, note that the x-axis is the change in debt to income, by County, from 2002 to 2006. And the y-axis is for the period 2006 - 2008. This shows that the change in the debt to income ratio was a good predictor of default rates.

From the authors:

Correlation across Counties of Default Rates and House Prices during Recession with Leverage Growth from 2002 t0 2006

The left panel presents the correlation across U.S. counties of the increase in the household debt to income ratio from 2002 to 2006 and the increase in the default rate from 2006 to 2008. The right panel presents the correlation across U.S. counties of the increase in the household debt to income ratio from 2002 to 2006 and the decline in house prices from 2006 to 2008. The sample includes 450 counties with at least 50,000 households as of 2000.

The second graph is figure 6A from the paper. From the authors:

Auto Sales and Unemployment Rates in High and Low Leverage Growth Counties

High leverage growth counties are defined to be the top 10% of counties by the increase in the debt to income ratio from 2002 to 2006. Low leverage growth counties are in the bottom 10% on the same measure. The left panel plots the growth in auto sales for high and low leverage growth counties since 2005, and the right panel plots the change in the unemployment rate in high and low leverage growth counties since 2005. Auto sales decline and unemployment increases by significantly more in counties that experience the sharpest increase in debt to income ratios from 2002 to 2006.

The third graph is figure 6B from the paper. From the authors:

Correlation across Counties of Auto Sales and Unemployment during Recession with Leverage Growth from 2002 to 2006

The left panel presents the correlation across U.S. counties of the increase in the household debt to income ratio from 2002 to 2006 and the decline in auto sales from 2006 to 2008. The right panel presents the correlation across U.S. counties of the increase in the household debt to income ratio from 2002 to 2006 and the increase in unemployment rates from 2006 to 2008. The sample includes 450 counties with at least 50,000 households as of 2000.

This analysis, comparing high and low leverage counties, is very revealing and shows that the high leverage areas are also the hardest hit (not surprising to those of us who felt mortgage equity extraction was a significant driver of consumption growth). The authors conclude:

[T]he initial economic slowdown was a direct result of an over-leveraged household sector unable to keep pace with its debt obligations.