This research explores the impact of CEOs' cognitive biases stemming from earnings management practices and financial distress on their future investment decisions. Existing literature shows a link between earnings management and excessive investment (McNichols and Stubben, 2008), and the impact of financial stress on investment choices (Bassetto and Kalatzis, 2011; López Gutiérrez et al., 2015). This study uniquely contributes by sequentially blending prospect theory, framing, conservatism, and psychological projection to explain how CEOs' optimism based on past earnings information affects future investment decisions. It also investigates how financial distress or soundness moderates the relationship between earnings management and future over/underinvestment decisions, further differentiating this association across developed and developing countries due to varying economic factors like economic rents, environmental uncertainties, and regulatory environments.

The study examines the intersection of earnings management and financial distress, noting how managers manipulate earnings to avoid bankruptcy or meet debt covenants (Rosner, 2003; Saleh and Ahmed, 2005). It also explores CEOs' cognitive biases, arguing that earnings management creates cognitive dissonance, leading CEOs to use earnings management to resolve this dissonance and maintain reputation (Festinger, 1957; Jiraporn et al., 2008). The role of firms' financial conditions (distress or soundness) in shaping risk aversion and framing is also considered, leading to potential psychological projections as self-defense mechanisms (Kahneman and Tversky, 1979; Kominis and Dudau, 2018). Prior research (Healy and Wahlen, 1999; McNichols and Stubben, 2008; Biddle et al., 2009; Lenard and Yu, 2012) informs hypotheses on the relationships between earnings management, financial conditions, and over/underinvestment decisions.

The study uses a sample of manufacturing firms listed on stock exchanges in 46 developed and developing countries (selected based on UN criteria and the Best Countries 2020 list) from 2009-2019, resulting in 70,587 firm-year observations after filtering for data completeness. Abnormal investment is measured as the difference between actual and predicted investment (Richardson, 2006; Baker et al., 2016), using a regression model incorporating growth, leverage, cash, firm age, size, and past returns. Earnings management is measured using discretionary accruals calculated with a modified Jones model (Dechow and Dichev, 2002; Guay et al., 1996; Kothari et al., 2005). Financial condition is measured by a modified Altman Z-score comparing actual and predicted liabilities (Pindado et al., 2008; Sanz and Ayca, 2006). Hypotheses are tested using regression analysis. A two-stage least squares (2SLS) approach is used to examine the intervening effect of financial conditions on the relationship between earnings management and investment decisions. The study includes control variables for firm size, leverage, profit growth, asset growth, profit margin, market-to-book ratio, and cash flow from operations.

The analysis reveals that earnings management with past income increases positively affects future overinvestment, while past income decreases positively affect future underinvestment. Firms’ financial distress positively impacts future overinvestment, and financial soundness positively affects future underinvestment. The interaction effect between financial conditions and earnings management on investment decisions was not significant, suggesting a sequential (rather than simultaneous) relationship. Using 2SLS analysis, the study confirms the intervening effect of financial distress on the relationship between income-increasing earnings management and overinvestment, and a similar intervening effect of financial soundness on the relationship between income-decreasing earnings management and underinvestment. These findings hold consistently for both developed and developing countries.

The findings support the hypothesis that CEOs' decisions regarding over/underinvestment are driven by past performance and financial pressures. The lack of significant moderating effects points to a sequential relationship where financial conditions first influence earnings management, which then affects investment decisions. The consistent results across developed and developing countries suggest a universalistic behavior pattern among CEOs, possibly due to the cognitive biases and self-defense mechanisms outlined in the framework. The negative framing caused by financial pressures results in heuristic behaviors and psychological projections in an attempt to overcome past failures and maintain reputation. These findings highlight the risk of abnormal investment decisions driven by cognitive biases and the importance of transparency to protect investors and creditors.

This study demonstrates the existence and mechanisms of CEO over- and under-investment, linked to earnings management and financial conditions. CEOs' cognitive biases, particularly the negative framing effect, create heuristic behavior and self-defense mechanisms, leading to risky investment choices. This phenomenon appears universal across developed and developing countries, emphasizing the need for transparency to protect investors. Future research should focus on more precise measurement of heuristic and self-defense mechanisms, including investigation into the impact of CEO tenure and a more comprehensive model that considers the sequential relationship among earnings management, investment decisions, and earnings acceleration, along with their impact on stock prices.

The study's limitations include the lack of direct measurement of heuristic and self-defense mechanisms and the exclusion of CEO tenure as a variable. Future research could improve on this by developing better instrumental variables for these mechanisms and including CEO tenure to examine its effects on the observed relationships. Further, the model could be extended to use a three-stage least squares (3SLS) approach that incorporates earnings acceleration and stock price changes to provide a more complete picture of the dynamics at play.