ARE251/Econ270A Plan for Second Half of Fall 2008

• Distribute Assignment 3. Idea: Have every student get a different LSMS survey, and use this in connection with lectures.

• Bardhan-Udry Ch. 2 (Household Economics)
• Deaton-Muellbauer 1980

Idea is to outline basic model, a la Singh et al via Bardhan & Udry. Show that separation is implied by basic model. Discuss how to estimate demand system, labor supply, all in an environment with no risk.

• Discussion of "Research Proposal"
  • Scope (confined to topics covered in class)
  • Form (Look at format for NSF proposals)

• Benjamin (1992)
• Deaton (2008)

Idea: What can go wrong? If there's risk, separation fails. Ways to test separation. Discuss concrete methods for testing for separation using a LSMS dataset.

• Draft of Ligon (2008)
• Bardhan-Udry Ch. 8
• Fafchamps (1992)

Idea: Effects of risk; measurement of risk. Models of full risk-sharing; interpretation of multipliers; relationship to interest rates.

• Townsend (1995)
• Kurosaki (1999)

Idea: How to test for full risk sharing.

• Browning, Bourguignon, Chiappori (2008)

Idea: Cover basic static model with no uncertainty

• Assignment 3 Due

Idea: Cover intra-household problem with time and uncertainty.

• Rangel-Thomas (2005)
• Duflo-Udry (2004)
• Lundberg-Pollak (1996)
• Mazzocco (2007)

• Jacoby-Mansuri (2008) REStud
• Dubois from the /New Palgrave's/ on Sharecropping

• Distribute Assignment 4

• Welfare measurement. Be constructive: start with ideal of measuring dynamic, intertemporal utility, and then introduce poverty, inequality, etc. as compromises that may sometimes be necessary.

Idea: Ideally, we could just observe everyone's preferences, assets, etc. Adopt a notion of cardinal utility, and estimate discounted expected utility.

• Old exams

• Deaton (Analysis of Household Surveys) Ch. 3
• Atkinson (1970)
• Harsanyi (1955)

Idea: Since observing everything is too hard, see what we can do with good panel data on household expenditures.

• Ligon (2005)
• Ligon-Schechter (2003)

• Assignment 4 Due

• Deaton Ch. 6
• Deaton-Paxson (1994)
• Ligon (2008)

Idea: Inequality is the best we can do with just cross-sectional data. With repeated cross-sections, we can construct artificial cohorts; with additional structure from theory on evolution of distribution, can do Deaton-Paxson or Ligon (2008) sorts of exercises.

• Research proposal due

• Foster-Greer-Thorbecke (1984)
• Ravallion (1998?) On comparing poverty and inequality measures.
• Chen-Ravallion (2007)

Without decent cross-sectional data (perhaps under-reporting at the top end), poverty measures may be a reasonably robust way to get at welfare. Or not.