Rational Inattention: A Solution to the Forward Discount Puzzle

Abstract

The uncovered interest rate parity equation is the cornerstone of most models in international macro.However, this equation does not hold empirically since the forward discount, or interest rate differential, is negatively related to the subsequent change in the exchange rate. This forward discount puzzle is one of the most extensively researched areas in international finance. It implies that excess returns on foreign currency investments are predictable. In this paper we propose a new explanation for this puzzle based on rational inattention. We develop a model where investors face a cost of collecting and processing information. Investors with low information processing costs trade actively, while other investors are inattentive and trade infrequently. We calibrate the model to the data and show that (i) inattention can account for most of the observed predictability of excess returns in the foreign exchange market, (ii) the benefit from frequent trading is relatively small so that few investors choose to be attentive, (iii) average expectational errors about future exchange rates are predictable in a way consistent with survey data for market participants, and (iv) the model can account for the puzzle of delayed overshooting of the exchange rate in response to interest rate shocks.

Investor attention and FX market volatility

“…Rational inattention slows down the process whereby new information becomes impounded into the exchange rate, leading to predictable excess returns. Bacchetta and Van Wincoop (2005) show that rational inattention provides a solution to the forward discount puzzle.…”

“…Only a small fraction of international financial holdings are actively managed (Sager and Taylor, 2006;Bacchetta and Van Wincoop, 2010). The infrequency of portfolio allocation decisions may be explained by optimal attention allocation, when information acquisition costs are added to transaction costs (Bacchetta and Van Wincoop, 2005). Rational inattention slows down the process whereby new information becomes impounded into the exchange rate, leading to predictable excess returns.…”

Figuring Figures: Exploring Europeans’ Knowledge of Official Economic Statistics

“…Moreover, costs might vary among individuals, being substantial for those who either do not know how and where to obtain information or do not have the skills to process and understand it (Sims 2003;Blinder and Krueger 2004). Therefore, it is postulated that, for some individuals, the costs might exceed the perceived benefits derived from information; consequently, they would rationally choose not to update (Mankiw and Reis 2002;Sims 2003;Bacchetta and van Wincoop 2005;Reis 2006). Additionally, some authors have hypothesized that individuals' demand for information might differ depending on the sources checked, the volume of news and their type (Akerlof et al 2000;Souleles 2001;Carroll 2003;Blinder and Krueger 2004;Curtin 2009).…”

Monetary Policy Shifts and the Forward Discount Puzzle

“…Bacchetta and Wincoop (2005) and Van Wincoop (2007, 2010) demonstrate that it could be optimal for rational decision makers to manage their portfolio only on occasion, as corresponding fees may be higher than gains from trading more frequently, leading to violations of the UIRP. But if this were the entire solution, a few powerful traders could take advantage of forward excess returns in the long run.…”