FREE ELECTRONIC LIBRARY - Thesis, dissertations, books

Pages:   || 2 | 3 | 4 | 5 |   ...   | 6 |

«Atif Mian Princeton University and NBER Kamalesh Rao MasterCard Advisors Amir Sufi University of Chicago Booth School of Business and NBER February ...»

-- [ Page 1 ] --

Household Balance Sheets, Consumption, and the Economic Slump

Atif Mian

Princeton University and NBER

Kamalesh Rao

MasterCard Advisors

Amir Sufi

University of Chicago Booth School of Business and NBER

February 2013


We show that the 2007-09 housing collapse in the United States resulted in a very unequal

distribution of wealth shocks due to the geographical concentration of ex-ante leverage and

house price decline. We investigate the consumption consequences of these wealth shocks and show that the consumption risk-sharing hypothesis is easily rejected. We estimate an elasticity of consumption with respect to housing net worth of 0.6 to 0.8 and an average marginal propensity to consume (MPC) of 5 to 7 cents for every dollar loss in housing wealth. However, the MPC is sharply higher for poorer and more levered households. Our findings thus highlight the role of debt and geographical distribution of wealth shocks in explaining the large and unequal decline in consumption from 2006 to 2009.

* Lucy Hu, Ernest Liu, Christian Martinez, Yoshio Nozawa, and Calvin Zhang provided superb research assistance.

We are grateful to the National Science Foundation, the Initiative on Global Markets at Chicago Booth, and the Fama-Miller Center at Chicago Booth for funding. Seminar participants at Chicago Booth, Columbia Business School, Harvard, MIT Sloan, MIT Economics, NYU Stern, Stanford GSB, UC Berkeley, UCLA, UC San Diego, and the NBER Monetary Economics meeting provided valuable feedback. The results or views expressed in this study are those of the authors and do not reflect those of the providers of the data used in this analysis.

Corresponding authors: Mian: (609) 258-6718, atif@princeton.edu; Sufi: (773) 702 6148, amir.sufi@chicagobooth.edu How does consumption respond to large negative shocks to household wealth? Do households with different levels of wealth have different marginal propensities to consume out of a dollar lost? These questions are fundamental in macroeconomics and finance, and the answers have profound implications for how we model the economy, how wealth shocks translate into business cycle fluctuations, and how policy should respond when asset prices collapse.

For example, most traditional models of the macro-economy adopt a representative agent framework, implicitly assuming that individual households are hedged against idiosyncratic or household-specific wealth shocks. However, if this assumption is grossly violated in data, then we may need to adopt heterogeneity in our models. Further, if households across the wealth distribution do not have the same marginal propensity to consume out of changes in wealth, then the distribution of dollar losses across the economy may matter for consumption dynamics.

These questions are especially important when considering severe recessions. In the United States, both the Great Depression and Great Recession were preceded by a large accumulation of debt and followed by a collapse in asset prices and consumption. 1 Recent theoretical research inspired by the Great Recession has focused on possible heterogeneity in marginal propensity to consume as an explanation for the large decline in spending. The heterogeneity in MPC is driven by differences in wealth, leverage and liquidity-access across households.

This paper provides detailed empirical evidence on the distribution of wealth shocks across the U.S. population at the onset of the Great Recession and on the consumption consequences of these wealth shocks. We put together a new data set that enables us to observe changes in household consumption and wealth at the county and zip code levels.

See for example Temin (1976) and Olney (1999) for evidence on the Great Depression. For the Great Recession, NIPA and Census retail sales data show a definitive collapse in durable consumption even before the fall of 2008.

We begin by documenting the large cross-sectional dispersion in changes in household wealth due to the collapse in housing market. Neighborhoods that accumulated a high level of debt during the housing boom were more likely to experience a fall in house prices between 2006 and 2009. The combination of debt and house price decline created huge losses in these neighborhoods. At the same time, areas that avoided accumulation of debt during the housing boom remained largely unscathed. The large cross-sectional difference in leverage build up and house price dynamics are in turn driven by differences in the terrain-based housing supply elasticity to a large extent.

We analyze how household spending responded to the large fall in household wealth. If households have sufficient mechanisms to insure their consumption against wealth shocks, as implicitly assume by a representative agent model, then we should see little to no response of consumption to wealth shocks. However, we find a very large elasticity of consumption with respect to the drop in housing net worth of between 0.6 and 0.8. We discuss why this estimate is unlikely to be driven by unobserved permanent income shocks.

Why do households cut consumption in response to idiosyncratic wealth shocks? We show that tightened credit constraints are partially responsible. In particular, households with larger decline in housing net wealth experience a stronger reduction in credit limit and greater difficulty in refinancing their mortgage into lower interest rates.

A second useful representation of the response of consumption to housing wealth shock is in terms of the marginal propensity to consume (MPC). Using the cross-sectional variation in consumption and net housing wealth decline, we estimate that consumption falls by between 5 and 7 cents for every dollar fall in housing net wealth.

A key question for the macroeconomic consequences of wealth shocks is whether there is heterogeneity across households in their MPC. In particular, a given decline in housing wealth is disproportionately borne by households that have an equity claim on the housing market.

Households that have a debt claim on housing are naturally protected, especially if the debt is insured by the government. If the MPC is the same for all, then it does not matter how wealth losses are distributed across various stake holders.

However, if MPC is (say) higher for borrowers with a levered equity claim on the housing market then the aggregate consumption consequences of housing wealth decline will be more severe the more levered the housing sector is. A unique advantage of have micro-level data on consumption, household balance sheet and house prices is that we can test for heterogeneity in MPC. We find that the MPC out of housing net wealth is much higher for poorer households, households with higher leverage and households that are more likely to be underwater.

For example, households with annual income less than $35 thousand have an MPC that is three times as large as the MPC for households with more than $200 thousand in income.

Similarly, households in the 90th percentile of the leverage distribution have an MPC that is twice as high as households in the 10th percentile of the leverage distribution. The heterogeneity in MPC is strongest in terms of the likelihood of being underwater.

This paper is related to the growing literature on understanding the role that household debt plays in generating severe business cycles. Cross-country business cycle studies by IMF (2012) and Jordà, Schularick and Taylor (2011) show that the presence of a high level of household debt leads to deeper recessions. Our paper is the first document the channel through which this might happen.

Our paper is also related to the vast literature in consumption theory and its empirical counterpart. We discuss some of this work in the next section. The next section also relates our work to some of the recent theoretical work on how financial shocks might generate deep and prolonged recessions. The remainder of our paper is structured as follows. Section 2 presents the data and summary statistics. Section 3 discusses variation in net worth shocks across counties.

Sections 4 and 5 present the results, and Section 6 concludes.

1. Theory How does a severe shock to net worth – like the collapse of house prices in the United States during the Great Recession – impact consumption and the real economy? Consider an = + + −

economy where households i’s net wealth at time t is given by:

–  –  –

The first three terms on the right hand side represent the market values of stocks, bonds, and housing, respectively, while the last term represents the value of debt borrowed by the household.

Imagine a severe negative shock to wealth unexpectedly strikes the economy. The wealth shock changes asset prices in the economy, which results in a change in household i's net worth.

Given the household's initial asset holdings, we can compute the change in household net worth − −1 = ∆ ∗ −1 + ∆ ∗ −1 + ∆ℎ ∗ −1 (in dollars) by:

–  –  –

respectively. Throughout, we use the symbol ∆ for growth, or percent change, in a variable. The debt term disappears from equation (2) because we are assuming that the value of debt is fixed in nominal terms, which implicitly disallows default, additional levering, or paying down debt 2. In equation (2), we focus on the change in net worth in dollar units, but we can define the change in

–  –  –

of equation (2) through by the lagged value of net worth for this household.

How should household consumption respond to the wealth shock? There is a large literature on this question, and we outline the basic hypotheses below.

A. The complete risk-sharing hypothesis Suppose households in the economy have CRRA preferences. Then, under the assumption of complete risk-sharing across households, growth in consumption is completely unrelated to idiosyncratic wealth shocks (e.g., Cochrane (1991)). In particular, any cross = + ∗ ∆ +

sectional regression relaying consumption growth to net worth growth of the form:

–  –  –

Equation (3) is derived under the strong assumption of complete markets. However, as Constantinides and Duffie (1996) discuss, under some restrictions on the income process, this relationship can also be derived with incomplete markets and limited borrowing capacity as long as people can trade in a few basic securities (see papers such as Telmer (1993) and Heaton and Lucas (1992, 1996)). Allowing for governmental transfer programs and informal insurance mechanisms provides yet another rationale for consumption insurance.

house prices, and hence for to be close to zero. Housing differs from other assets because it is There is one more reason why households are naturally hedged against movements in also a consumption good. As a result, for a homeowner who expects to live in his home for a Our empirical results are robust to factoring in the effect of default on net wealth loss. See appendix for details.

long time or who cares about his offspring to live in a similar home, an increase in house prices does not make him richer because it also increases the implicit rental cost of housing. A similar argument works when house prices decline. Under this view, households should not be responsive to movements in net worth driven by home values. 3 A corollary of the above argument is that a reduction in house prices may increase nonhousing consumption for households that were planning on increasing their housing consumption in the future. An example would be renters planning on buying a bigger condominium or home in the future. They actually feel richer when house prices decline. Homeowners that were planning on downsizing may decrease their non-housing consumption as they now feel poorer in real terms. In the aggregate, these offsetting effects would lead to a diminished effect of housing net worth on consumption.

One advantage of our empirical approach is that the data are aggregated at the zip code or county level. The data therefore aggregate consumption information for homeowners and renters within a zip code or county. Moreover, the correlation between the homeownership rate and the net worth shock due to the collapse in house prices during the Great Recession at the zip code our empirical estimate of incorporates the net effect of responses by homeowners and renters.

level is statistically indistinguishable from zero at the 1% and 5% confidence level. As a result, The theoretical argument that consumption should be unresponsive to movement in house prices depends on households having standard preferences, rational asset prices, and no credit market frictions. 4 However, in a world where housing serves as collateral as well, the riskSee Campbell and Cocco (2007) for this argument. Sinai and Souleles (2005) make the additional point that home ownership provides a hedge against future fluctuations in rental cost.

The literature on the housing wealth effect is too large to be completely summarized here--it includes Muellbauer and Murphy (1997), Attanasio and Weber (1994), Lehnert (2004), Case, Quigley, and Shiller (2005 and 2013), Haurin and Rosenthal (2006), Campbell and Cocco (2007), Greenspan and Kennedy (2007), and Bostic, Gabriel, and Painter (2009), and Carroll, Otsuka, and Slacalek (2011)).

sharing prediction of = 0 may not hold. We return to this argument in the empirical work below.

macroeconomics. If in equation (3) were indeed close to zero, then households would be The complete risk-sharing hypothesis plays a crucial role in finance and hedged against household-specific wealth shocks, and we would not need to track households separately. Instead, a single “representative agent” would provide a sufficient description of the entire macro-economy and idiosyncratic wealth shocks would play no role in explaining the Given the theoretical importance of equation (3), a number of studies estimate in the cross-section of consumption growth.

Pages:   || 2 | 3 | 4 | 5 |   ...   | 6 |

Similar works:

«Chapter 1 EVOLUTIONARY DECISION TREES FOR STOCK INDEX OPTIONS AND FUTURES ARBITRAGE How Not To Miss Opportunities Sheri Markose Economics Department and Institute of Studies in Finance University of Essex scher@essex.ac.uk Edward Tsang Computer Science Department University of Essex edward@essex.ac.uk Hakan Er Economics Department University of Essex her@essex.ac.uk Abstract EDDIE-ARB (EDDIE stands for Evolutionary Dynamic Data Investment Evaluator) is a genetic program (GP) that implements a...»

«THE OFFICIAL BUSINESS CREDIT BUILDING CHECKLIST Your Checklist to Building Business Credit Easily, Quickly & Effortlessly The Official Business Credit Building Checklist The Official Business Credit Building Checklist E very highly successful business in the United States has business credit established. Many of the largest businesses in the world such as Walmart even used their business credit to get as big as they are. As essential as business credit is for a company to have real success,...»

«Consumable Worksheets Strand 3 Skill 6 The other curriculum has the health that a additional plan with a post by the number card. Anticipate inward of this hassles that they are rush lots largely routed to their pdf whether store and communication mobi. You can have cold exclusions of they do coffee of 504 and more residual minutes. The family Consumable Worksheets Strand 3 Skill 6 with process customer pdf has pdf setting is for your circumstances are to plan. One amongst the most other...»

«381 CONSUMING THE PUBLIC SCHOOL David F. Labaree School of Education Stanford University Abstract. In this essay David Labaree examines the tension between two competing visions of the purposes of education that have shaped American public schools. From one perspective, we have seen schooling as a way to preserve and promote public aims, such as keeping the faith, shoring up the republic, or promoting economic growth. From the other perspective, we have seen schooling as a way to advance the...»

«BANK FOR INTERNATIONAL SETTLEMENTS IMPLICATIONS FOR CENTRAL BANKS OF THE DEVELOPMENT OF ELECTRONIC MONEY Basle October 1996 © Bank for International Settlements 1996. All rights reserved. Brief excerpts may be reproduced or translated provided the source is stated. ISBN 92-9131-059-X BANK FOR INTERNATIONAL SETTLEMENTS IMPLICATIONS FOR CENTRAL BANKS OF THE DEVELOPMENT OF ELECTRONIC MONEY Basle October 1996 FOREWORD Since the end of 1995 the central banks of the Group of Ten (G-10) countries...»

«TD/B/C.I/MEM.7/11 United Nations United Nations Conference Distr.: General 5 August 2015 on Trade and Development Original: English Trade and Development Board Trade and Development Commission Multi-year Expert Meeting on Transport, Trade Logistics and Trade Facilitation Fourth session Geneva, 14–16 October 2015 Item 3 of the provisional agenda Sustainable freight transport systems: Opportunities for developing countries Note by the UNCTAD secretariat Executive Summary The year 2015 is a...»

«European Journal of Business and Social Sciences, Vol. 2, No.7, pp 78-104, October 2013. P.P. 78 104 URL: http://www.ejbss.com/recent.aspx ISSN: 2235 -767X VENTURE CAPITAL: SURVEY OF THE ECONOMIC IMPACT AND THE DETERMINANTS EMPIRICAL EVIDENCE FROM EGYPT Dr. Elsayed Elsiefy Associate Professor, QFISHamad Bin Khalifa UniversityQatar Foundation, Faculty of CommerceAlexandria University, Egypt. Visiting Scholar, Islamic Legal Studies Program, Harvard Law School. Harvard University ABSTRACT he...»

«Enterprise Risk Management and the Cost of Capital Thomas R. Berry-Stö a lzle Jianren Xub,* April 30, 2014 * Please address all correspondence to Jianren Xu. a Terry College of Business, University of Georgia, 206 Brooks Hall, Athens, GA 30602, Tel.: +1-706-542-5160, Fax: +1-706-542-4295, trbs@uga.edu b Terry College of Business, University of Georgia, 206 Brooks Hall, Athens, GA 30602, Tel.: +1-706-542-1721, Fax: +1-706-542-4295, jxu@uga.edu Enterprise Risk Management and the Cost of Capital...»

«Dear Tourism Partner We are delighted to welcome you to Ireland this week and we look forward to meeting with you and to showcasing what Ireland has to offer as a compelling and competitive holiday destination. We are very excited about our immediate business prospects. Our industry has made tremendous progress in enhancing our competitive position internationally. Also, visitor sentiment, tracked independently, is particularly strong and interest in Ireland amongst the international travel...»

«Date: January 6, 2006 Author: Ken Landoline RESEARCH UPDATE FrontRange GoldMine IP Voice Suite Software Delivers Cost Effective Enhanced Functionality to Enterprise Knowledge Workers Summary FrontRange Solutions, Inc. (FrontRange), a leader in small to medium business and mid-market enterprise software development and provider of the well-known HEAT® service management and GoldMine® Customer Relationship Management (CRM) product suites, has announced a product that launches the company into...»

«Curriculum Vitae Josef Baumgartner Research staff member Personal information First name(s), surname Josef Baumgartner Address Austrian Institute of Economic Research 1030 Vienna  Austria, Arsenal, Objekt 20 Telephone +43 1 798 26 01-230 Fax +43 1 798 93 86 E-mail Josef.Baumgartner@wifo.ac.at Homepage http://josef.baumgartner.wifo.ac.at/ Date of birth 6 December 1964 Nationality Austria Research area(s) Macroeconomics and European Economic Policy Scope of work Business cycle, inflation and...»

«Data Processing Agreement for Oracle Cloud Services Version December 1, 2013 1. Scope and order of precedence This is an agreement concerning the Processing of Personal Data as part of Oracle’s Cloud Services (“Cloud Services”), as further specified in (i) the applicable Oracle master agreement and (ii) the Oracle Cloud Ordering Document between Customer and Oracle, and all documents, addenda, schedules and exhibits incorporated therein (collectively the “Agreement”) by and between...»

<<  HOME   |    CONTACTS
2016 www.dis.xlibx.info - Thesis, dissertations, books

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.