« Back to Results

New Approaches in Measuring Uncertainty

Paper Session

Saturday, Jan. 6, 2018 12:30 PM - 2:15 PM

Loews Philadelphia, Washington C
Hosted By: Society for Economic Dynamics
  • Chair: David William Berger, Northwestern University

Firm Expectations: Measuring Subjective Uncertainty

Nicholas Bloom
,
Stanford University
Steven Davis
,
University of Chicago
Lucia Foster
,
U.S. Census Bureau
Brian Lucking
,
Stanford University
Scott Ohlmacher
,
University of Maryland
Itay Saporta
,
Tel Aviv University

Abstract

An innovative new Census survey asked 5-bin outcome and probability questions on around 35,000 US manufacturing establishments over future expectations of sales, employment, investment and materials. This paper will present new results from this survey. Using similar data sets from the Atlanta Fed and Bank of England we find that, first, over 90% of respondees can provide consistent responses to the questions, suggesting most managers can answer detailed expectation questions. Firms unable to respond usually have lower productivity and growth rates suggesting non-response is indicative of lower performance. Second, the first three moments of expectations – the means, variance and skewness of growth rates - of sales, employment and investment are consistent with historical values, suggesting expectations are well founded. Finally, expectations of uncertainty are strongly correlated with historic growth volatility of firm and industry level outcomes, suggesting these volatility and dispersion measures provide excellent proxies of firm uncertainty.

Firms’ Uncertainty and Ambiguity

Ruediger Bachmann
,
University of Notre Dame
Kai Carstensen
,
University of Kiel
Martin Schneider
,
Stanford University

Abstract

Businesses deal with uncertainty every day. Hiring, investment and pricing decisions are difficult in part because future demand, cost, competition and regulation cannot be perfectly foreseen. While economists have long studied the effects of uncertainty on business decisions in theory, measurement of those effects is in its infancy. This project tackles this challenge head on by asking firms in the IFO’s Business Cycle Survey what their views on future sales growth as well as assessments of their likelihood are and this new survey is one of the first anywhere to elicit perceptions of Knightian uncertainty in the field. It provides crucial data to test hypotheses on behavior under such uncertainty as well as allowing us to measure the real effects of uncertainty and ambiguity shocks.

Technological Innovation and the Distribution of Labor Income Growth

Leonid Kogan
,
Massachusetts Institute of Technology
Dimitris Papanikolaou
,
Northwestern University
Lawrence Schmidt
,
University of Chicago
Jae Song
,
Social Security Administration

Abstract

We examine how the distribution of worker earnings growth shifts following major
technological advances by the firm, or its competitors. We combine the firm-level
measure of the value of innovation developed in Kogan, Papanikolaou, Seru, and
Stoffman (2016) with administrative employer-employee matched data. We find that
technological innovation is associated with an increase in the dispersion of labor income growth, even for the firm's own workers. We find an economically significant increase in the left tail of income growth for the highest paid workers in innovating firms. The empirical results are consistent with the view that innovation displaces part of workers' human capital. In particular, we find that the increase in the left tail is larger for older relative to younger workers; is driven by process innovation; and mostly reflects the income dynamics of workers that leave the firm rather than stay.

Uncertainty Shocks as Second-moment News Shocks

David William Berger
,
Northwestern University
Ian Dew-Becker
,
Northwestern University
Stefano Giglio
,
University of Chicago

Abstract

This paper provides new empirical evidence on the relationship between aggregate uncertainty and the macroeconomy. We identify uncertainty shocks using methods from the literature on news shocks, following the observation that second-moment news is a shock to uncertainty. The key distinction we draw is between realized volatility { the realization of large shocks { and forward-looking uncertainty { the expectation that future shocks will be large. According to a wide range of VAR specifications, shocks to realized stock market volatility are contractionary, while shocks to uncertainty have no significant effect on the economy. In line with those findings, investors have historically paid large premia to hedge shocks to realized volatility, but the premia associated with shocks to uncertainty have not been statistically different from zero. We argue that these facts, and the VAR identification, are consistent with a simple model in which output growth is skewed left. Aggregate volatility matters, but it is the realization of volatility, rather than uncertainty about the future, that seems to be associated with declines.
JEL Classifications
  • D2 - Production and Organizations
  • G1 - General Financial Markets