IQ, Expectations, and Choice

Abstract

Forecast errors for inflation decline monotonically with both verbal and quantitative IQ in a large and representative male population. Within individuals, inflation expectations and perceptions are autocorrelated only for men above the median by IQ (high-IQ men). High-IQ men's forecast revisions are consistent with the diagnostic-expectations framework, whereas anything goes for low-IQ men. Education levels, income, socioeconomic status, or financial constraints do not explain these results. Using ad-hoc tasks in a controlled environment, we investigate the channels behind these results. Low-IQ individuals' knowledge of the concept of inflation is low; they associate inflation with concrete goods and services instead of abstract economic concepts, and are less capable of forecasting mean-reverting processes. Differences in expectations formation by IQ feed into choice-only high-IQ men plan to spend more when expecting higher inflation as the consumer Euler equation prescribes. Our results have implications for heterogeneous-beliefs models of consumption, saving, and investment.

Household Inflation Expectations and Consumer Spending: Evidence from Panel Data

“…In contrast, Duca, Kenny, and Reuter (2019) use cross-sectional survey data from the euro area to document that higher expected inflation boosts readiness to spend in a ZLB environment. D'Acunto et al (2019) find that among Finnish men, only high-IQ individuals act in a manner that indicates a positive relationship between inflation expectations and the readiness to purchase durables. Two related papers on German and Polish consumers Weber 2016 and2018, respectively) find that an unexpected announcement of an increase in the consumption tax boosts inflation expectations and readiness to spend on durables by economically significant amounts.…”

“…The differential response between college-educated and other respondents may reflect differences in IQ, a factor that, as noted above, is found to mediate the planned spending response to inflation expectations (D'Acunto et al 2019). The robust response of college-educated mortgagors suggests such households understand that higher inflation erodes their real mortgage obligations.…”

Expectations with Endogenous Information Acquisition: An Experimental Investigation

“…In contrast, updating rates do not differ by respondent education (not shown) 27. The lower slope in the follow-up is expected, because individuals may have been exposed to additional signals during the interim four months, thus gradually diluting the effect of the signal provided during our experiment 28. We winsorize this variable at the 1.5th percentile (0.46 minutes) and at the 98.5th percentile (18.79 minutes) of the distribution.…”

“…Column (2) of Table 2 uses the log time spent on the screen used to report the posterior belief as the dependent variable. 28 Due to the design of the survey, this variable includes the time spent looking at the information, and therefore respondents who get to see information would mechanically take more time reporting the posterior belief.…”

Managing Households' Expectations with Salient Economic Policies

“…Taken together, our results show that large differences exist in the effectiveness of policies that operate theoretically through the same channels but differ in the salience of their implications to consumers. Consumers' non-response to forward-guidance announcements, either in terms of inflation expectations or consumption plans, suggests that limited cognitive 9 See, e.g., D'Acunto, Malmendier, Ospina, and Weber (2018); D'Acunto, D'Acunto, Hoang, Paloviita, and Weber (2019c)).…”