Production and Operations Management, 2025, 34(6), 1256–1275

How Do Companies Change Their IT Spending After a Data Breach?

Data breaches are among the most costly threats facing businesses today. But how do companies react after their systems are hacked? Our study explores a psychological dilemma managers face: do they ‘learn from failure’ by investing boldly in better systems, or do they fall into ‘threat rigidity’, becoming paralysed by fear and cutting back on new investments?

By analysing data from US public firms over a seven-year period, we discovered that companies actually do both. After a data breach, firms tend to significantly increase their overall IT investment intensity — spending more money and deploying more resources to fix vulnerabilities, restore customer trust, and strengthen security. At the same time, they drastically reduce investments in new, innovative IT solutions. Because new technologies carry unknown risks, managers become highly conservative in a crisis, preferring to stick with what they know rather than experimenting with unproven technologies.

Key findings:

• Investment intensity: After a data breach, firms channel more resources into existing IT systems, focusing on repairs, upgrades, and customer reassurance.
• New investment caution: At the same time, firms cut back on new technologies, reflecting managers’ reluctance to take risks during periods of uncertainty.
• Cultural influence: Firms with strong quality cultures are more likely to boost IT spending to fix the root causes of a data breach. Conversely, firms with strong innovation cultures are less afraid of the unknown and are less likely to freeze investments in new technologies.

Ultimately, our study reveals a clear managerial implication: a firm's reaction to a data breach is a tale of two strategies. Managers must carefully balance the need to strengthen security with the caution required when adopting new innovations.

Operations Research, 2026, Online publication

Integrating Descriptive and Prescriptive Analytics: Optimised Dimensionality Reduction

Moment based distributionally robust optimisation (DRO) provides a framework for making decisions under conditions of uncertainty. Its strength lies in accounting for worst-case scenarios by combining statistical information with optimisation techniques. The challenge is that DRO problems often involve very high-dimensional data, which makes them computationally demanding and slow to solve.

Our study focuses on multiproduct newsvendor problems and production transportation planning — classic operations challenges where uncertainty about demand or logistics can make solutions time-consuming. Most DRO problems can be reformulated as semidefinite programming (SDP) problems, which are solvable in theory, but high-dimensional SDPs remain difficult in practice.

We introduce an optimised dimensionality reduction (ODR) approach that integrates the reduction of random parameters directly into the optimisation process. Unlike existing methods that reduce dimensions first and then solve the simplified problem, our integrated approach produces low-dimensional models that are both faster to solve and closer to the optimal solution.

Key findings:

• Efficiency gains: The ODR approach reduces computational time dramatically —up to three orders of magnitude — while still delivering solutions that are optimal or nearly optimal.
• Accuracy improvement: By integrating dimensionality reduction with optimisation, our method achieves results within 0.1% of the true optimal value, outperforming leading benchmark approaches.
• Practical applications: Our approach proves effective in multiproduct newsvendor and production-transportation problems, showing clear advantages over traditional methods.

The implication is that complex optimisation problems under conditions of uncertainty can now be solved with far greater speed and accuracy. By adopting ODR, organisations can handle large-scale, uncertain scenarios more efficiently, ensuring robust solutions without being slowed down by computational bottlenecks.

Journal of Marketing, 2026, Online publication

Influencer marketing has become a mainstream way for brands to reach consumers. Selection decisions often rely on visible metrics such as follower counts, likes, and audience demographics, together with a sense of ‘fit’. This study highlights an overlooked but important factor: an influencer’s endorsement rate, defined as the proportion of brand-sponsored posts relative to their total posts.

The core insight is that endorsement rate doesn’t follow a simple rule — neither “a higher rate is always better” nor “a lower rate is always better”. Instead, engagement with sponsored posts tends to be strongest at the extremes and weaker in the middle.

Based on two field studies and two experiments on platforms including Instagram, Twitter, and Douyin, we find that engagement with sponsored posts varies with endorsement rate. Engagement tends to be stronger when endorsement rates are either low or high, but weaker when they are moderate. This pattern arises because endorsement rate triggers two opposing consumer reactions. A higher rate can signal brand recognition (“this creator is sought after by brands; they must be influential”). However, frequent sponsorship can also raise suspicion of manipulative intent (“they’re paid to say this; can I trust it?”). Engagement reflects the balance of these forces.

Key findings:

1. Engagement pattern: Endorsement rate follows a U-shaped curve, with stronger engagement at low and high levels, and weaker engagement at moderate levels.
2. Authenticity cues: Organic product mentions (non-sponsored references) and consistent brand endorsements (repeated partnerships with the same brand) can soften scepticism, making feeds feel more genuine and reducing the negative impact of high endorsement rates. In other words, how ‘commercial’ a feed looks matters, but so does whether the influencer appears genuinely interested in products and maintains coherent, ongoing brand relationships rather than one-off promotions.

A real-world scenario illustrates the managerial value. Imagine a skincare brand choosing between two creators with similar follower counts. Creator A posts sponsored content occasionally; Creator B posts sponsored content very frequently. A manager may reflexively prefer Creator A, fearing that Creator B looks too ‘salesy’. Our study suggests a more nuanced approach: Creator B may still generate strong engagement if audiences interpret the high endorsement rate as a signal of status — especially if Creator B also posts authentic, organic skincare routines and maintains consistent long-term partnerships.

The takeaway is to treat endorsement rate as a strategic screening and planning variable. Managers should select creators with an eye for where they sit on the U-curve, and design content calendars and partnerships that build credibility while minimising persuasion scepticism.

Accounting Review, 2026, 101 (1): 411–435

In this study, we examine the impact of a shared common auditor among suppliers and customers on trade credit. Using hand-collected pairwise trade credit data, we find that a supplier extends more trade credit to a customer audited by a common auditor. This association is robust to alternative design specifications and various sample restrictions to alleviate selection bias. We then conduct interviews with trade credit managers and executives as a prelude to archival analyses exploring multiple potential mechanisms to explain this association. The collective results are most consistent with the explanation that mutual third parties to a dyadic relationship can foster trust through social connections and increased salience of reputation effects rather than the explanation that a common auditor reduces information asymmetry about the rigor of the audit process.

Journal of Financial and Quantitative Analysis, Forthcoming

We show that the decision to go public is influenced by spatial variation in the supply of equity financing. We measure the amount of capital of equity investors in each US region and document that the incidence of initial public offerings (IPOs) by intangible-intensive resident firms increases significantly when regional equity capital is abundant. Using a novel empirical strategy and hand-collected data on out-of-state pension flows, we confirm that our findings are not due to underlying regional factors.

Contemporary Accounting Research, Forthcoming

In this study, we examine how achievement-related tendencies are expressed in the professional auditing context, particularly through the interplay between the CEO and the audit partner. We use the facial width-to-height ratio (fWHR), a stable morphological trait widely applied in prior research, as a proxy for achievement drive. Using a sample of US audit partners from 2016 to 2019, we find that higher achievement drive is associated with enhanced audit quality, evidenced by fewer restatements and lower abnormal accruals. Auditors with higher achievement drive are also more likely to become industry experts, attain leadership positions, and achieve partnership status more quickly. Importantly, we find that high-achievement-drive audit partners are more inclined to assert dominance in negotiations, particularly when working with equally driven CEOs, leading to improved audit quality. Overall, our findings suggest that, when activated in auditing contexts, achievement-oriented tendencies, as proxied by fWHR, are linked to higher audit quality.

Problem definition: New product introductions frequently face retail imitation risks. When start-ups use crowdfunding to finance new products, they expose themselves to an even more severe imitation risk specific to crowdfunding platforms. To mitigate this risk, start-ups can employ a progressive launch strategy by initially offering a preliminary version with fewer features or functionalities. Alternatively, they may opt for traditional bank financing, which only carries retail imitation risk. This paper examines how start-ups can effectively compete with crowdfunding copycats through the aforementioned product launch and financing strategies. Methodology/results: We develop a game-theoretical model to analyze the strategic interactions between a start-up and a copycat. Our findings reveal that start-ups may strategically withhold product quality provision to deter or weaken crowdfunding copycats. The optimal quality provision in crowdfunding exhibits nonmonotonic relationships with both crowdfunding imitation efficiency and crowdfunding market share. We also evaluate the social impact of different financing strategies and reveal that the socially optimal funding choice depends on the implementation of the progressive launch strategy. Notably, contrary to conventional wisdom, increasing crowdfunding copycat efficiency does not consistently improve consumer surplus or social welfare. Managerial implications: Although crowdfunding offers numerous advantages, start-ups considering this approach must recognize and address the potential threats posed by copycat imitation. Our research indicates that start-ups seeking external financing can effectively use progressive launch strategies to counter crowdfunding copycats—an approach that generally requires less time and resources than traditional intellectual property protection methods.