The Discomfort of Doing Nothing: Why Non-Intervention Outperforms Action Across Domains, and Why We Resist It Anyway

Abstract

Across strikingly different domains, from personal finance to medicine to parenting to mental health, a similar pattern recurs: restraint frequently outperforms intervention, yet people and institutions consistently intervene anyway. Passive investors reliably outperform active traders and most professional fund managers (Barber & Odean, 2000; Fama & French, 2010). Many medical conditions are better managed through monitoring than through immediate treatment, and unnecessary diagnosis and treatment carry their own documented harms (Welch et al., 2011). Overinvolved, or "helicopter," parenting is associated with worse psychological outcomes in children than more hands-off approaches (Schiffrin et al., 2014). And a meaningful proportion of people with major depression recover without any treatment at all (Whiteford et al., 2013). This essay argues that a single well-documented psychological mechanism, action bias, the tendency to prefer doing something over doing nothing even when inaction produces better outcomes, helps explain why non-intervention is so hard to choose and sustain (Patt & Zeckhauser, 2000; Bar-Eli et al., 2007). The essay also identifies the conditions under which this pattern reverses, since restraint is not universally superior, and closes with practical implications for individuals and institutions who want to treat "doing nothing" as a legitimate option rather than a failure to act.

The Discomfort of Doing Nothing: Why Non-Intervention Outperforms Action Across Domains, and Why We Resist It Anyway

A pattern shows up across remarkably unrelated fields. Investors who trade frequently, trying to time the market and pick winners, underperform investors who simply buy and hold a diversified portfolio. Patients who receive aggressive, immediate treatment for many conditions often do no better, and sometimes worse, than patients who are carefully monitored instead. Children whose parents intervene constantly in their problems tend to show worse psychological adjustment than children given more room to struggle and self-correct. And people experiencing a depressive episode sometimes recover on their own, without therapy or medication, at rates that complicate simple stories about what "caused" a given recovery. In each case, the instinct to act, to trade, to treat, to intervene, to fix, is understandable and often well-intentioned. It is also, on average and across many documented cases, less effective than waiting. This essay examines four bodies of evidence for that pattern and then turns to the psychological research that explains why doing nothing is so difficult to choose, even when it is demonstrably the better option, before addressing where the pattern breaks down.

Investing: The Well-Documented Cost of Active Trading

The clearest evidence for the superiority of restraint comes from finance. Barber and Odean (2000) analyzed the trading records of tens of thousands of households at a large discount brokerage and found that the more frequently investors traded, the worse their net returns became; the households in the highest-turnover quintile earned annual returns several percentage points lower than those in the lowest-turnover quintile, a gap driven almost entirely by trading costs and poor security selection rather than by bad luck. Their now-famous conclusion, that trading is hazardous to your wealth, was not a claim about a few unlucky traders; it described a systematic pattern in which the specific stocks investors bought underperformed the ones they had just sold, meaning that the act of trading itself, not merely its costs, was destroying value.

This pattern extends to professionals, not just amateurs. Fama and French (2010) examined the entire cross-section of actively managed U.S. equity mutual funds and found that, once the costs of active management were accounted for, the aggregate portfolio of active funds performed no better than the market as a whole, and few individual funds generated returns sufficient to cover their own costs. In other words, the average professional fund manager, paid specifically to identify winning trades and outmaneuver the market, could not reliably do so once fees were included. The passive alternative, a low-cost index fund that makes no active bets at all, has for decades been the option that both academic finance and, increasingly, financial advisors recommend as a default, precisely because it removes the costly temptation to act.

Healthcare: When Watching Beats Treating

A parallel argument has emerged in medicine, most prominently in Welch et al.'s (2011) analysis of overdiagnosis, the detection and labeling of abnormalities that were never going to cause a patient harm in their lifetime. Welch and colleagues document how many cancers detected through aggressive screening are slow-growing and would never have become symptomatic, yet once detected, the diagnosis itself creates enormous pressure to treat, exposing patients to the real risks of surgery, radiation, or chemotherapy for a condition that active surveillance, monitoring the condition closely without immediately intervening, would have managed just as well. This logic underlies the now-standard practice of active surveillance for certain low-risk prostate cancers and "watchful waiting" for a range of other conditions, in which a clinician deliberately chooses to observe rather than treat, based on evidence that early, aggressive intervention would expose the patient to greater harm than benefit. The pattern is not limited to cancer; it recurs any time a condition is common, slow-moving, and heterogeneous in how much it will actually affect a given patient's life.

Parenting: Overinvolvement and Child Well-Being

A similar pattern appears in child development research. Schiffrin et al. (2014) surveyed college students on the degree to which their parents engaged in "helicopter parenting,", overcontrolling involvement that persists past the developmental point at which it is age-appropriate, and found that students with more overcontrolling parents reported significantly higher depression and lower life satisfaction. Critically, the authors traced this effect to a specific mechanism: excessive parental intervention undermined students' sense of autonomy and competence, the basic psychological needs that, according to self-determination theory, must be satisfied for a person to function well. In other words, the parental intervention was not simply unhelpful; it appeared to actively interfere with the developmental process it was intended to support, by not giving children the room to encounter, tolerate, and resolve manageable problems on their own.

Mental Health: Recovery Without Treatment

The pattern reaches even into the treatment of mental illness itself. Whiteford et al. (2013) conducted a systematic review and meta-analysis of studies tracking people with major depression who had not received treatment, and found a meaningful proportion of these cases remitted within a year even without any intervention, with remission rates varying considerably by symptom severity and duration. This finding does not imply that depression is not a real or serious condition, nor that treatment is ineffective; severity moderated the results considerably, and the studies included in this literature are themselves limited by the ethical impossibility of randomly assigning severely depressed patients to remain untreated. What it does establish is that a person's improvement following a period without treatment cannot automatically be attributed to the passage of time being harmless, since for a substantial subset of cases, time alone was doing the work that is often credited to intervention when it does happen to coincide with treatment.

A Common Mechanism: Action Bias

Why, given all of this evidence, do people and institutions continue to intervene? Behavioral decision research offers a specific answer: action bias, the well-documented tendency to prefer action over inaction even when there is no rational reason to expect action to produce a better outcome (Patt & Zeckhauser, 2000). Patt and Zeckhauser demonstrated this experimentally using hypothetical environmental decisions: when given a choice between actively improving one site or passively preserving another, decision-makers systematically favored the option that let them "achieve" a visible improvement, even when the framing made the choice of which site to help arbitrary, and even when action required imposing losses elsewhere. The bias was strong enough that participants sometimes chose to actively make things worse rather than simply do nothing.

A vivid real-world illustration of the same bias comes from professional sports. Bar-Eli et al. (2007) analyzed hundreds of penalty kicks in elite soccer and found that goalkeepers dive to the left or right the overwhelming majority of the time, even though the statistically optimal strategy, staying in the center of the goal, produces a meaningfully higher save rate. The authors' explanation is instructive: because the social norm is to dive, a goal scored while the goalkeeper stood still feels worse, both to the goalkeeper and to observers, than a goal scored while diving in the wrong direction. Inaction that fails is judged more harshly than action that fails, even when the action was less likely to succeed. This asymmetry in how failure is judged, not evidence about what actually works, appears to drive the choice.

This single mechanism plausibly explains the pattern across all four domains discussed above. An investor who loses money after actively trading feels they at least "did something"; an investor who loses money while sitting still feels they failed to act when they should have. A physician who treats a slow-growing cancer and the patient later has complications is protected by having followed the instinct to intervene; a physician who recommends watchful waiting and the condition later worsens faces harder questions, regardless of what the statistics say about the average patient. A parent who intervenes in a child's problem and the child struggles anyway is judged as caring; a parent who deliberately steps back is at greater risk of being seen as neglectful, even if the child would have developed better coping skills. In each case, the visible, effortful choice is socially and psychologically safer than the invisible, restrained one, independent of which choice actually produces the better outcome.

Where the Pattern Reverses

None of this evidence supports the conclusion that inaction is always superior, and the same studies that document the cost of overintervention also document where intervention is genuinely necessary. Severity is the recurring boundary condition. Welch et al. (2011) do not argue against treating symptomatic disease, only against treating conditions that would never become symptomatic. Whiteford et al. (2013) found that spontaneous remission rates were considerably lower and slower among the most severely depressed patients, meaning the population at greatest risk is precisely the population for whom "waiting it out" is least supported by the evidence. Schiffrin et al. (2014) explicitly note that some parental involvement supports positive child outcomes; the harm was specifically associated with overinvolvement that persisted past developmental appropriateness, not with parental engagement as such. And even in finance, Fama and French (2010) found evidence of genuine skill in the extreme tails of the fund distribution, meaning some active management does add value, even though it is not reliably identifiable in advance. The lesson is not "never act," but that the decision to act should be treated as a genuine choice requiring justification, rather than a default that inaction must argue its way out of.

Implications

If action bias is the common mechanism behind unnecessary intervention across these domains, the most direct countermeasure is a structural one: deliberately building "do nothing, and monitor" into the set of options considered before a decision is made, rather than only after intervention has already been attempted and failed. In investing, this looks like a written plan committing to a passive strategy before market volatility creates the emotional pull to trade. In medicine, it looks like formal active-surveillance protocols with defined monitoring intervals, so that watchful waiting is a structured clinical pathway rather than an anxiety-inducing absence of care. In parenting, it looks like explicitly identifying which problems a child is developmentally capable of handling alone before stepping in. In mental health treatment, it looks like distinguishing, as far as the evidence allows, between presentations likely to remit with time and support alone and those where the severity and trajectory clearly warrant active treatment, and being honest with patients about that distinction rather than defaulting to intervention as the only legible form of care. In each case, the goal is not to eliminate action, but to make restraint a visible, defensible, and socially acceptable choice, so that it no longer has to compete on an uneven playing field against the instinct to simply do something.

Conclusion

The evidence reviewed here spans finance, medicine, parenting, and mental health, four domains with almost nothing in common except this: in each, a substantial body of peer-reviewed research shows that restraint frequently outperforms intervention, and in each, people continue to intervene anyway. Action bias offers a parsimonious explanation for why this gap persists. It is not that people are irrational or poorly informed; it is that inaction which fails is judged, and felt, differently than action which fails, regardless of the underlying probabilities. Recognizing this bias does not mean treating "do nothing" as always correct. It means treating it as a real option, worth the same scrutiny, evidence, and structural support that intervention currently receives by default.

References

Bar-Eli, M., Azar, O. H., Ritov, I., Keidar-Levin, Y., & Schein, G. (2007). Action bias among elite soccer goalkeepers: The case of penalty kicks. Journal of Economic Psychology, 28(5), 606–621. https://doi.org/10.1016/j.joep.2006.12.001

Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of individual investors. The Journal of Finance, 55(2), 773–806. https://doi.org/10.1111/0022-1082.00226

Fama, E. F., & French, K. R. (2010). Luck versus skill in the cross-section of mutual fund returns. The Journal of Finance, 65(5), 1915–1947. https://doi.org/10.1111/j.1540-6261.2010.01598.x

Patt, A., & Zeckhauser, R. (2000). Action bias and environmental decisions. Journal of Risk and Uncertainty, 21(1), 45–72. https://doi.org/10.1023/A:1026517309871

Schiffrin, H. H., Liss, M., Miles-McLean, H., Geary, K. A., Erchull, M. J., & Tashner, T. (2014). Helping or hovering? The effects of helicopter parenting on college students' well-being. Journal of Child and Family Studies, 23(3), 548–557. https://doi.org/10.1007/s10826-013-9716-3

Welch, H. G., Schwartz, L. M., & Woloshin, S. (2011). Overdiagnosed: Making people sick in the pursuit of health. Beacon Press.

Whiteford, H. A., Harris, M. G., McKeon, G., Baxter, A., Pennell, C., Barendregt, J. J., & Wang, J. (2013). Estimating remission from untreated major depression: A systematic review and meta-analysis. Psychological Medicine, 43(8), 1569–1585. https://doi.org/10.1017/S0033291712001717