Subjective Expectations and Income Processes in Rural India

Abstract

This paper uses unique primary data on directly elicited individual subjective expectations to analyse and characterize the process that generates the income of poor, rural Indian households. We validate and use responses to subjective expectations questions and a parametric assumption to fit a household‐specific probability distribution for future income. Combining computed moments from this distribution with data for actual current income, we specify and estimate a dynamic model of household income. We find that our households face a very persistent income process. Our paper is one of the first that uses subjective expectations data to model income processes.

Subjective Expectations of Medical Expenditures and Insurance in Rural Ethiopia

“…In total, 177 respondents in 2012 (11.5%) and 255 in 2013 (16.4%) reported probabilities that are logically inconsistent. This is higher than the 4 percent rate of logical response errors found by Attanasio and Augsburg (2012) in their study of subjective income expectations in rural India using a similar instrument but is comparable to Dominitz and Manski's (1997) finding of a 10 percent error incidence for subjective expectations of income in Wisconsin. 5 Similar to those excluded because of non-response or enumerator errors, households providing illogical responses are more likely to be poorer, smaller, less healthy and more likely to forgo health care for economic reasons (Appendix Table A1).…”

“…This proceeds by first asking preliminary questions to fix the range of the distribution of future expenditure and then asking the respondent to report the probability of exceeding thresholds within this range (Dominitz and Manski 1997;Manski 2004;Attanasio and Augsburg 2012). The specific questions used to determine the minimum and maximum amounts were: The reference to a new illness is intended to ensure that long-standing health conditions that will continue to require treatment and medication are not overlooked in reporting the minimum and that the maximum refers to expenses that would be incurred were health to deteriorate markedly.…”

“…The household-specific mean of forecast annual expenditure (i.e. the first moment of the subjective distribution) ranges from ETB 2.40 (US$ 0.14) to ETB 6375 (US$ 365.97), with a sample mean of ETB 461 (US$ 6 From only three subjective probabilities it is not possible to establish which type of distribution fits the data best (Attanasio and Augsburg, 2012).…”

Subjective Expectations of Medical Expenditures and Insurance in Rural Ethiopia

“…In total, 177 respondents in 2012 (11.5%) and 255 in 2013 (16.4%) reported probabilities that are logically inconsistent. This is higher than the 4 percent rate of logical response errors found by Attanasio and Augsburg (2012) in their study of subjective income expectations in rural India using a similar instrument but is comparable to Dominitz and Manski's (1997) finding of a 10 percent error incidence for subjective expectations of income in Wisconsin.…”

“…This proceeds by first asking preliminary questions to fix the range of the distribution of future expenditure and then asking the respondent to report the probability of exceeding thresholds within this range (Dominitz and Manski 1997;Manski 2004;Attanasio and Augsburg 2012). The specific questions used to determine the minimum and maximum amounts were:…”