Home » Research Blog » Information inattentiveness of financial markets

Information inattentiveness of financial markets

Academic research explains macroeconomic information inefficiency with “stickiness” and “signal extraction problems”. Information stickiness means that forecasts cannot be updated continuously and hence markets partly operate on outdated information. Signal extraction problem means that forecasters struggle to separate noise from signal in economic data. The consequence is rational “inattentiveness” of financial markets, offering profit opportunities to those that analyze economic data timely and efficiently.

Easaw, Joshy, and Roberto Golinelli (2014), “Inflation Expectations and the Two Forms of Inattentiveness”, Cardiff Economics Working Papers, No. E2014/21, October 2014

Mankiw, Gregory and Ricardo Reis (2002), “Sticky Information Versus Sticky Prices: A Proposal to Replace the New Keynesian Phillips Curve”, The Quarterly Journal of Economics, November 2002

Sims, Christopher (2003), “Implications of rational inattention”, Journal of Monetary Economics 50, pp/ 665–690

Coibion, Olivier, and Yuriy Gorodnichenko (2012), “Information Rigidity and the Expectations Formation Process: A Simple Framework and New Facts”, IMF Working Paper 12/296, December 2012

The below are excerpts from various papers. Emphasis and cursive text have been added.

Popular models of information inefficiency

“Until recently most macroeconomic models simply assumed that [professional] forecasts were formed rationally with full information…[However, now] different forms of information rigidities, or agent’s inattentiveness, form the basis of the competing rational expectations models with informational frictions:

  • There is the sticky-information model…Here the agents update their information set sporadically. Agents do not continuously update their expectations but choose an optimal time at which to be inattentive, that is they receive no news about the economy until it is time to plan again.…
  • Informational friction models argue that agents update their information set continuously but can never fully observe the true state because of signal extraction problems…”

[Easaw and Golinelli]

Understanding sticky information

“The essence of the [sticky-information] model is that information about macroeconomic conditions diffuses slowly through the population. This slow diffusion could arise because of either costs of acquiring information or costs of re-optimization. In either case, although prices are always changing, pricing decisions are not always based on current information…In this sticky-information model…expectations are…important, but the relevant expectations are past expectations of current economic conditions.”

[Mankiw and Reis]

“The degree of information rigidity in [the sticky information] model is then the probability of not acquiring new information each period.”

[Coibion and Gorodnichenko]

Understanding signal extraction problems

“Agents continuously update their information sets…via a signal extraction problem. Forecasts are a weighted average of agents’ prior beliefs and the new information received, where the weight on prior beliefs can be interpreted as the degree of information rigidity.”

[Coibion and Gorodnichenko]

“That people have limited information-processing capacity should not be controversial. It accords with ordinary experience…The…information-processing constraint…arises from coding inaccuracy, the need to approximate fine-grained information with cruder approximations. People share a need to code macroeconomic data efficiently, and they pay for this service. To a considerable extent, people’s needs for coding are similar, so they rely on common sources of coded information. The result is that a considerable part of the erratic response arising from information capacity constraints is common across individuals.”


“Consider one of the most pervasive coding services, the daily newspaper, and how it presents macroeconomic data. Many newspapers report the Federal Funds Rate to three significant figures every day, at a predictable location in the back of the business section. The vast majority of newspaper readers do not look at this number every day, and of those that do look at the page containing the number, the vast majority make no adjustment in their behavior in reaction to the number. On the other hand, if the New York Times ran as a threecolumn, front page headline ‘‘FED UNEXPECTEDLY RAISES FEDERAL FUNDS RATE 1.5%,’’ many readers of the newspaper would be likely to act on the news.”


Evidence for information inattentiveness

“We consider two forms of inattentiveness…

  • The standard inattentiveness arises when the agents try to revise or update their expectations formed in the previous period [as there will always be some share of forecasts that are not fully updated and hence ‘old’ numbers are being used].
  • We also consider an additional form of inattentiveness. Typically, in each period, professional forecasters not only revise their forecast from the previous period but also form multi-period forecasts. The second form of inattentiveness arises when the agent is trying to distinguish the forecasts between the different horizons [as there will always be some confusion as to what in the data is trend and what is short-term distortion]…

Both instances involve the ability to observe relevant but different information and, therefore, the forms of inattentiveness… the first form of inattentiveness is a necessary condition for the second form to exist…the existence of inattentiveness when revising their forecasts necessitates resorting to their multi-period forecasts in the previous period…The empirical investigation using various surveys of professional forecasts for both the US and the UK establishes the existence of both forms of inattentiveness.”

[Easaw and Golinelli]

We document widespread rejections of full-information rational expectations in exactly the direction predicted by models of information rigidities. Consistent with these models, when one takes into account forecast revisions, other macroeconomic variables lose much of their ability to predict forecast errors.”

[Coibion and Gorodnichenko]


Related articles