Huge If True: Learning from the Size of the Surprise

arXiv:2611.15747 · econ.GN, q-fin.EC · Submitted 2026-11-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.

Ruth: We're looking at the paper: "Huge If True: Learning from the Size of the Surprise".

Warren: Here's the gist of the paper. Responses to earnings surprises depend on an unknown multiplier mapping earnings to value, which investors learn about from the surprise itself.

Ruth: This multiplier uncertainty leads to an S-shaped response: greater uncertainty about this multiplier attenuates responses on average but makes them steeper near zero surprises, meaning investors respond more strongly to small surprises than large ones.

Warren: The central observation is that the size of the surprise provides information about both cash flows and the multiplier, allowing investors to learn "the news and how much the news matters." This inference results in an S-shaped response function.

Ruth: The model predicts that greater multiplier uncertainty lowers average sensitivity while raising sensitivity to small surprises, distinguishing it from other explanations for the S-shape of price responses.

Paper summary: Ruth: So, to recap the core idea of "Huge If True: Learning from the Size of the Surprise," it centers on an unknown multiplier that connects earnings to value, and how learning about that multiplier through surprise magnitude shapes investor reactions.

Warren: The paper claims that this unknown multiplier uncertainty leads investors to respond less on average when uncertainty is high, but it makes responses steeper near zero surprises. This means they respond more strongly to small surprises than large ones.

Iris: They model this S-shaped response function, suggesting that greater multiplier uncertainty steepens the slope of the response curve right at the consensus point while flattening it off as you move away from that consensus.

Ruth: The authors argue that this is distinct from other explanations for S-shaped responses because it’s tied to how investors use the surprise size to infer information about both cash flows and that multiplier itself.

Warren: They test this by finding that greater multiplier uncertainty lowers the average sensitivity but raises sensitivity near the consensus, which is what distinguishes their finding from standard accounts of decision difficulty where harder problems weaken responses everywhere.

Iris: They use valuation problems as a source for this uncertainty because mapping news to value depends on unobservable quantities like future cash flow growth and firm risk, which can support very different valuations depending on investor expectations.

Ruth: The paper is pretty thorough in its testing, confirming that greater multiplier uncertainty is associated with weaker responses on average but stronger responses near the consensus across both experimental and field evidence.

Warren: It’s interesting because they found this effect holds up when they use valuation uncertainty as a proxy for the multiplier uncertainty, showing it's not just an artifact of one specific setup.

Iris: They also note that they didn't find evidence of underreaction or overreaction tied to valuation uncertainty, suggesting that the mechanism is about changing what news people focus on most strongly depending on how hard it is to map value from the earnings report.

Conclusion: Ruth: So, looking at "Huge If True: Learning from the Size of the Surprise," we see that this paper moves beyond just saying things get harder; it specifies *how* that difficulty affects investor learning about the earnings news.

Warren: The authors’ main point is that multiplier uncertainty doesn't just dilute reactions generally; it reshapes them, making investors hyper-sensitive to small surprises right around the consensus point.

Iris: They show this effect is robust, confirmed in both lab experiments and real-world data using valuation uncertainty as a measure of this multiplier uncertainty. This points to a very specific informational channel at play.

Ruth: What this implies for us is that when you're analyzing market reactions to earnings, especially those near the consensus, you have to consider the size of the surprise not just as an indicator of news magnitude, but as a signal about how much information investors gain about the underlying value mapping.

Warren: It suggests that if you're trying to build a trading strategy or model investor behavior, you can't treat decision difficulty as a blanket dampener; you need to account for this size-dependent sensitivity around the consensus.

Iris: The authors also highlight their limitation, which is that they are focusing on how uncertainty about the multiplier affects response, but they don't delve deeply into what exactly drives that uncertainty in the field settings beyond using valuation uncertainty as a proxy.

Ruth: That’s fair. They aren't fully mapping out every single source of that multiplier uncertainty in detail, but they’ve shown a very clear mechanism for how it affects the response function itself, which is what matters here.

Thomas Graeber, Christopher Roth, Marco Sammon, Constantin Schesch

University of Zurich · LMU Munich · Harvard Business School · Sciences Po

econ.GN, q-fin.EC

Submitted: 2026-11-11

Updated: 2026-11-13

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 100/100

The gist: Responses to earnings surprises depend on an unknown multiplier mapping earnings to value, which investors learn about from the surprise itself.

Key concepts

Multiplier Uncertainty
This refers to the unknown relationship between a company's earnings and its market value. Investors learn about this multiplier directly from how much an earnings surprise impacts the stock price, creating uncertainty about how much any given news will matter.
S-shaped Response Function
This describes the pattern of investor reactions to surprises: responses are weaker on average but become steeper near zero surprises (small changes) and flatter for large surprises. This shape is a key finding explained by multiplier uncertainty.
Processing Constraints
This refers to limitations or difficulties investors face when trying to process information, such as how much effort they put into analyzing a surprise. Manipulating these constraints helps isolate the effect of uncertainty about the earnings-to-value multiplier.

Terminology

Summary

Responses to earnings surprises depend on an unknown multiplier mapping earnings to value, which investors learn about from the surprise itself. p. 0

Where the paper says it

Huge If True: Learning from the Size of the Surprise∗ Thomas Graeber Christopher Roth Marco Sammon Constantin Schesch November 11, 2026 Abstract Responses to earnings surprises depend on an unknown multiplier mapping earnings to value. Investors, however, learn about the news and the multiplier from the same number.

This multiplier uncertainty leads to an S-shaped response: greater uncertainty about this multiplier attenuates responses on average but makes them steeper near zero surprises, meaning investors respond more strongly to small surprises than large ones. p. 0
Where the paper says it

Large surprises may indicate small multipliers and tiny surprises large ones. We model how greater uncertainty about the multiplier attenuates responses on average but makes them steeper near zero surprises.

The central observation is that the size of the surprise provides information about both cash flows and the multiplier, allowing investors to learn the news and how much the news matters. This inference results in an S-shaped response function.

The model predicts that greater multiplier uncertainty lowers average sensitivity while raising sensitivity to small surprises, distinguishing it from other explanations for the S-shape of price responses. p. 6

Where the paper says it

However, we place uncertainty on the multiplier that maps an observed signal into the appropriate response, which allows for a behavioral interpretation of the mechanism in terms of processing constraints. Since S-shaped responses to news can arise from many sources, however, we view the shape itself as only weak evidence for any particular mechanism, and instead derive and test a comparative static that this literature has not formulated or tested: greater multiplier uncertainty lowers average sensitivity while raising sensitivity to small surprises.

Empirical tests confirm this:

  1. An experiment manipulating processing constraints showed that tighter constraints reduced sensitivity over the full range by about one fourth while increasing it around the consensus by about one fourth. p. 3

    Where the paper says it

    Under tighter processing constraints, respondents react less strongly to large earnings surprises but more strongly to small ones, which makes the response function more S-shaped. In a simple piecewise-linear specification, tighter constraints reduce sensitivity over the full range of earnings news by about one fourth, while increasing sensitivity around the consensus by about one fourth.

  2. Field data using valuation uncertainty (VU) as a proxy for multiplier uncertainty showed that high-VU firms have a steeper return response around the analyst consensus and a flatter response for larger surprises. p. 4

    Where the paper says it

    Empirically, we find that valuation uncertainty is associated with the same change in the shape of the response function predicted by the model. Specifically, high-VU firms have a steeper return response around the analyst consensus and a flatter response for larger surprises.

The paper finds that greater multiplier uncertainty steepens the response near the consensus while flattening it farther away, which is complementary to other explanations for the S-shape. p. 7

Where the paper says it

We view these forces as determinants of the unconditional earnings-response function and ask instead how uncertainty about the earnings-to-value mapping changes its shape. Our model predicts that greater multiplier uncertainty steepens the response around the consensus while flattening it farther away: accounting for the non-linear nature of the response is therefore critical to test this comparative static both in the experiment and in the field.

The model suggests investors react optimally given their information, and there is no evidence of post-announcement continuation or reversal associated with valuation uncertainty. p. 5
Where the paper says it

Our model, on the other hand, suggests that investors react optimally given the information they have, so it does not imply post-announcement return predictability. Empirically, we do not find evidence of continuation or reversal.

The key comparative static is that greater multiplier uncertainty attenuates responses on average but steepens the slope of the response function near the consensus.

The authors test two sources of multiplier uncertainty: randomly tightening processing constraints in an experiment and using valuation uncertainty (VU) and textual measures of valuation difficulty in field evidence. In both settings, greater multiplier uncertainty is associated with weaker responses to earnings news on average but stronger responses near the consensus. The findings are robust across different specifications and controls. Furthermore, they show that these findings are not driven by established explanations for the S-shaped response of prices to earnings news, concluding that only measures which directly speak to multiplier uncertainty appear to steepen the S-shape exhibited by the market return reaction.

The model predicts that greater multiplier uncertainty raises sensitivity near the prior mean while lowering sensitivity farther away from the mean, a prediction confirmed empirically. The paper concludes that there is no clear indication of VU-related underreaction or overreaction, suggesting that making a decision harder can change which news people respond to most strongly.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Enhance decision-making under uncertainty by incorporating an S-shaped response function into AI models to reflect how investors learn about both the magnitude and importance of news, leading to greater sensitivity to small surprises when processing constraints are tight.

  2. Develop a mechanism for inferring unknown mappings between inputs and outputs, allowing AI systems to adapt their risk assessment by learning from the size of the surprise as an indicator of multiplier uncertainty.

  3. Improve signal interpretation by creating a framework where extreme signals are treated as evidence of lower precision or lower precision, leading to more calibrated responses that focus on the sign rather than magnitude when uncertainty is high.

  4. Implement robust response functions in AI that show average attenuation but simultaneously demonstrate an increase in sensitivity around the consensus, distinguishing this from standard models where decision difficulty weakens responses everywhere.

  5. Create a comparative static for AI systems by testing whether they exhibit weaker responses to large surprises on average but stronger responses near the consensus, which is predicted by multiplier uncertainty rather than just general processing difficulty.

Abstract

Responses to earnings surprises depend on an unknown multiplier mapping earnings to value. Investors, however, learn about the news and the multiplier from the same number. Large surprises may indicate small multipliers and tiny surprises large ones. We model how greater uncertainty about the multiplier attenuates responses on average but makes them steeper near zero surprises. In an incentivized experiment, we find that greater processing constraints lower average sensitivity but raise it near zero. We confirm this with real-world earnings announcements: stocks with higher valuation uncertainty, our proxy for multiplier uncertainty, respond less on average but more steeply near zero.

Related papers