Does risk-averse demand drive the negative slant of news? Testing the McCluskey–Swinnen–Vandemoortele bad-news mechanism and its interaction with supply-side framing

McCluskey, Swinnen and Vandemoortele (2015) proposed a purely demand-side explanation for the pervasive negative slant of news coverage: when news is instrumentally valuable and consumers are risk averse, diminishing marginal utility makes information about potential losses more valuable than information about equal-sized gains, so profit-maximizing newspapers devote more space to bad news. The note itself left the mechanism untested against data and abstracted from newspapers' strategic framing of news and from non-instrumental consumption motives. The open problem is to establish whether the risk-aversion channel actually operates in news markets — as distinct from taste-based negativity bias, anticipatory-utility motives, and supply-side slant — and to characterize how it interacts with strategic framing by news producers. The model's sharp comparative-static predictions (the slant strengthens with the loss-value asymmetry and with consumer risk aversion, and reverses when positive shocks are sufficiently larger than negative ones) supply testable restrictions that distinguish this channel from taste-based accounts of negativity.

Background

News coverage across domains — crime, health risks, economic performance — devotes systematically more space to negative events than to positive ones. This regularity is often called the negative slant or negativity bias of news. It is documented for economic news by Soroka (2006), who shows that US and Canadian consumer sentiment responds more strongly to negative than to positive macroeconomic news, and by Heinz and Swinnen (2015), who find that negative economic news receives roughly a factor-20 more media coverage than positive news of comparable newsworthiness. Theoretical explanations available before 2015 were either demand-side — in the market-for-news model of Mullainathan and Shleifer (2005), competing newspapers slant their reports toward readers' prior beliefs — or supply-side, where Gentzkow and Shapiro (2006) show how newspapers' concern for reputation with ideologically heterogeneous readers can sustain bias; Gentzkow and Shapiro (2010) document empirically that consumer demand is a strong driver of the political slant of US newspapers. Gentzkow, Shapiro and Stone (2015) survey this marketplace-of-media-bias literature, organizing it into demand-side explanations (readers wanting belief-confirming news) and supply-side explanations (reputation, ideology, commercial pressure, media capture by interested parties).

McCluskey, Swinnen and Vandemoortele (2015) — the source of this problem — offered a demand-side mechanism that requires no psychological bias on the part of consumers. In their model, newspapers compete for readers by choosing the space $q_g^i$ and $q_b^i$ they devote to positive ("good") and negative ("bad") news. A risk-averse consumer with income $Y$ and increasing, strictly concave utility $u_k(\cdot)$ obtains an expected utility gain of $\rho(q_g^i)[u_k(Y+z_g)-u_k(Y)]$ from good news (which lets the consumer act to capture an income gain $z_g$) and $\rho(q_b^i)[u_k(Y)-u_k(Y-z_b)]$ from bad news (which lets the consumer act to avoid a loss $z_b$), where $\rho(\cdot)$, the probability that the reader can act on a story, is increasing and concave in the space devoted to it. Diminishing marginal utility implies $u_k(Y+z)-u_k(Y)<u_k(Y)-u_k(Y-z)$ for symmetric shocks $z_g=z_b=z$: information about losses is worth more to a risk-averse reader than information about gains of the same size. Profit maximization then implies that the equilibrium allocation satisfies $q_b^{i*}>q_g^{i*}$ — newspapers supply more bad news than good news even though consumers have no taste for negativity, purely because bad news is more instrumentally valuable for loss avoidance. The prediction is not merely qualitative: the slant strengthens with the value asymmetry between losses and gains and with the degree of utility concavity (i.e., with risk aversion), and it reverses when the positive shock is sufficiently larger than the negative one.

This instrumental mechanism is conceptually distinct from the taste-based negativity bias documented in psychology and communication research, and separating the two is the crux of the open problem. Trussler and Soroka (2014), in a lab study, found that participants choose to read negative news stories even though they state a preference for positive news. Soroka, Fournier and Nir (2019) measured physiological stress responses (skin conductance and heart rate) to real video news in a 17-country experiment and found stronger reactions to negative than positive content on average. Robertson et al. (2023) analyzed roughly 105,000 headline variations from Upworthy.com as randomized controlled trials (N = 22,743 tests, about 5.7 million clicks) and showed that each additional negative word in a headline increases the click-through rate — the share of viewers who click — by about 2.3%, while positive words decrease it. Soroka, Daku, Hiaeshutter-Rice, Guggenheim and Pasek (2018) document a negativity bias in economic news coverage in both traditional and social media, and Noh and Soroka (2025) review the online evidence, where the size and stability of negativity effects remain debated. These studies establish that consumers respond to negativity, but through affective and attentional channels; none measures consumer risk preferences or exploits the source model's distinguishing predictions. Related to the "mixed consumption motives" the source model abstracts from, Ganguly and Tasoff (2017) show that anticipatory utility — utility derived from beliefs about future outcomes, such as suspense and dread — substantially shapes individuals' demand for information about future gains and losses, but not within a media-market equilibrium.

Problem Statement

The source note establishes theoretically that risk-averse consumer demand alone can generate a negative slant in news coverage; the prediction hinges on the symmetry of income shocks and the degree of utility concavity, and the slant can reverse when the positive shock is sufficiently larger than the negative one. No empirical calibration of the mechanism is provided, and the analysis abstracts from newspapers' strategic framing of news and from additional consumption motives. The open problem is therefore: does the risk-aversion channel proposed by McCluskey, Swinnen and Vandemoortele (2015) actually operate in news markets, and how does it interact with the supply side?

The question has an empirical and a theoretical dimension. Empirically, whether the observed negative slant of news reflects risk-aversion-driven demand for loss-relevant information, as distinct from the rival explanations — taste-based negativity bias, anticipatory-utility motives, and supply-side slant or framing — remains open. The model's comparative statics — the slant varies with the size asymmetry between positive and negative income shocks (reversing when the positive shock is sufficiently larger) and with the degree of consumer risk aversion — offer distinguishing restrictions that separate instrumental loss-avoidance demand from taste-based negativity. Theoretically, the source model abstracts from newspapers' strategic framing of news and from non-instrumental consumption motives; characterizing the conditions under which the demand-driven negative slant survives, weakens, or reverses when newspapers also choose how to frame news (and may face asymmetric production costs between good and bad news) would clarify the interaction with the supply side, including the restrictions such an extension places on observable news content and consumption data. The single scientific objective uniting both dimensions is to establish whether risk-aversion-driven consumer demand is a real cause of negative news slant and to characterize its interaction with supply-side framing.

References

1. Jill J. McCluskey, Johan Swinnen and Thijs Vandemoortele (2015), 'You get what you want: A note on the economics of bad news', Information Economics and Policy 30:1–5, DOI 10.1016/j.infoecopol.2014.10.003, https://doi.org/10.1016/j.infoecopol.2014.10.003
2. Claire E. Robertson, Nicolas Pröllochs, Kaoru Schwarzenegger, Philip Pärnamets, Jay J. Van Bavel and Stefan Feuerriegel (2023), 'Negativity drives online news consumption', Nature Human Behaviour 7(5):812–822, DOI 10.1038/s41562-023-01538-4, https://doi.org/10.1038/s41562-023-01538-4
3. Marc Trussler and Stuart Soroka (2014), 'Consumer Demand for Cynical and Negative News Frames', The International Journal of Press/Politics 19(3):360–379, DOI 10.1177/1940161214524832, https://doi.org/10.1177/1940161214524832
4. Stuart N. Soroka, Patrick Fournier and Lilach Nir (2019), 'Cross-national evidence of a negativity bias in psychophysiological reactions to news', Proceedings of the National Academy of Sciences 116(38):18888–18892, DOI 10.1073/pnas.1908369116, https://doi.org/10.1073/pnas.1908369116
5. Stuart Soroka, Mark Daku, Dan Hiaeshutter-Rice, Lauren Guggenheim and Josh Pasek (2018), 'Negativity and Positivity Biases in Economic News Coverage: Traditional Versus Social Media', Communication Research 45(7):1078–1098, DOI 10.1177/0093650217725870, https://doi.org/10.1177/0093650217725870
6. Seonhye Noh and Stuart Soroka (2025), 'Negativity Biases Online', Journal of Media Psychology 37:296–306, DOI 10.1027/1864-1105/a000479, https://doi.org/10.1027/1864-1105/a000479
7. Matthias Heinz and Johan Swinnen (2015), 'Media slant in economic news: A factor 20', Economics Letters 132:18–20, DOI 10.1016/j.econlet.2015.04.011, https://doi.org/10.1016/j.econlet.2015.04.011
8. Matthew Gentzkow, Jesse M. Shapiro and Daniel F. Stone (2015), 'Media Bias in the Marketplace', in Handbook of Media Economics, vol. 1, pp. 623–645, DOI 10.1016/B978-0-444-63685-0.00014-0, https://doi.org/10.1016/B978-0-444-63685-0.00014-0
9. Jill J. McCluskey, Nicholas Kalaitzandonakes and Johan Swinnen (2016), 'Media Coverage, Public Perceptions, and Consumer Behavior: Insights from New Food Technologies', Annual Review of Resource Economics 8:467–486, DOI 10.1146/annurev-resource-100913-012630, https://doi.org/10.1146/annurev-resource-100913-012630
10. Sendhil Mullainathan and Andrei Shleifer (2005), 'The Market for News', American Economic Review 95:1031–1053, DOI 10.1257/0002828054825619, https://doi.org/10.1257/0002828054825619
11. Matthew Gentzkow and Jesse M. Shapiro (2006), 'Media Bias and Reputation', Journal of Political Economy 114:280–316, DOI 10.1086/499414, https://doi.org/10.1086/499414
12. Matthew Gentzkow and Jesse M. Shapiro (2010), 'What Drives Media Slant? Evidence From U.S. Daily Newspapers', Econometrica 78:35–71, DOI 10.3982/ecta7195, https://doi.org/10.3982/ecta7195
13. Ananda Ganguly and Joshua Tasoff (2017), 'Fantasy and Dread: The Demand for Information and the Consumption Utility of the Future', Management Science 63:4037–4060, DOI 10.1287/mnsc.2016.2550, https://doi.org/10.1287/mnsc.2016.2550
14. Stuart N. Soroka (2006), 'Good News and Bad News: Asymmetric Responses to Economic Information', The Journal of Politics 68:372–385, DOI 10.1111/j.1468-2508.2006.00413.x, https://doi.org/10.1111/j.1468-2508.2006.00413.x