Estimating the elasticity of intertemporal substitution: Is the aggregate financial return free from the weak instrument problem? Fábio Augusto Reis Gomes a , Lourenço S. Paz b,⇑ a FUCAPE Business School, Av. Fernando Ferrari, 1358, Vitória-ES, 29075-505, Brazil b Syracuse University, 110 Eggers Hall, Syracuse, NY 13244, USA article info Article history: Received 12 April 2012 Accepted 23 January 2013 Available online 8 February 2013 JEL classification: C22 C25 E21 Keywords: Asset returns Consumption Elasticity of intertemporal substitution Model specification Weak instruments abstract The elasticity of intertemporal substitution (EIS) is one of the key parameters in the Eco- nomics and Finance literature. It is usually estimated by means of the consumer’s Euler Equation using an instrumental variable approach, and the estimates are usually zero or close to zero. Nevertheless, such attempts present two major problems: first, the use of weak instruments, and second, the absence of a rate of return that is representative of the agent’s asset portfolio. The latter has been addressed by using the return of a synthetic mutual fund (SMF), which is a weighted combination of the returns of all assets held by the average household. The use of SMF returns led to EIS estimates of about 0.2 for the US econ- omy. In this paper, we first investigate whether the EIS estimates using the SMF returns for the US suffer from the weak instrument problem. Next, we conduct robustness analyses using different estimators and instrument sets. Our findings show that estimates using SMF returns are plagued by weak instruments, but in some cases partially robust estima- tors were able to deliver a positive and statistically significant EIS estimate. Furthermore, we found that the Treasury Bill return does not suffer from weak instruments, but the EIS is not precisely estimated and seems to be close to zero. Ó 2013 Elsevier Inc. All rights reserved. 1. Introduction For several years, researchers have struggled to obtain consistent estimates of the elasticity of intertemporal substitution (EIS). This parameter is crucial not only in Economics but also in Finance. For instance, in the consumption and portfolio choice problem discussed in Campbell and Viceira (1999), the EIS is the key parameter in the optimal consumption rule. The theoretical framework often used to estimate the EIS comes from the idea that consumers smooth consumption in order to maximize their lifetime utility. Nevertheless, as pointed out by Browning and Crossley (2001), the researcher needs to choose a particular setup in order to arrive at an empirical specification that can be estimated. The specification often used throughout the literature is depicted by Eq. (1), hereafter called ‘basic setup’, lnðC t Þ¼ a i þ wr i;t þ e i;t ; i ¼ 1; 2; ... ; N ð1Þ 0164-0704/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmacro.2013.01.005 ⇑ Corresponding author. Tel.: +1 315 443 5874; fax: +1 315 443 3717. E-mail address: lspaz@maxwell.syr.edu (L.S. Paz). Journal of Macroeconomics 36 (2013) 63–75 Contents lists available at SciVerse ScienceDirect Journal of Macroeconomics journal homepage: www.elsevier.com/locate/jmacro