In 2014, a survey by the U.S. Bureau of Labor Statistics revealed that roughly half of American workers were subject to some form of "pay secrecy policy" — whether through formal contractual provisions or informal workplace culture.[1] In Asia, the figure is likely even higher. A study of companies in Japan, South Korea, and Taiwan found that over 70% of respondents considered "inquiring about a colleague's salary to be inappropriate behavior."[2] Why has a seemingly simple piece of information — "how much do you earn per month" — become such a potent taboo? More importantly, what impact does the existence of this taboo have on the distribution of interests between employers and employees?

I. Information Asymmetry: The Invisible Advantage at the Bargaining Table

1.1 Akerlof's "Market for Lemons" and Its Lessons

George Akerlof, the 2001 Nobel laureate in economics, revealed in his seminal paper "The Market for Lemons" how information asymmetry distorts market function.[3] In the used car market, sellers know the true condition of their vehicles, but buyers do not. This information gap causes buyers to offer only the "average quality" price, which in turn drives good cars out of the market, leaving behind only "lemons" — defective vehicles.

Salary negotiations in the labor market exhibit a similar structure, but in the opposite direction: here, it is the employer who possesses more complete information. Employers know:

  • The salary levels and distribution of all employees within the company
  • The going market rate for comparable positions
  • The company's salary budget and room for flexibility
  • The salary expectations of other candidates

Employees, on the other hand, typically know only their own salary and have no basis for comparison. This information gap places employees at a structural disadvantage in negotiations.[4]

1.2 BATNA in Negotiation Theory

The Harvard Negotiation Project introduced a core concept known as "BATNA" (Best Alternative to a Negotiated Agreement), which illustrates the critical role of information in negotiations.[5] Put simply: your bargaining power depends on how good your "walk-away" option is.

In salary negotiations, if an employee is unaware of market rates and colleagues' compensation, they cannot accurately assess their own BATNA. They may overvalue the company's offer (because they do not realize they are being underpaid) or underestimate the cost of leaving (because they are unaware of external opportunities). In either case, a lack of information leads to suboptimal decision-making.

Conversely, employers — armed with complete information — can precisely set salaries at the "lowest point an employee is barely willing to accept," known in economics as the "reservation wage."[6] This means the distribution of labor surplus is systematically tilted in favor of employers.

1.3 Mathematical Model: Quantifying the Value of Information

We can illustrate the value of information using a simplified mathematical model.

Suppose the "true value" of a position (the maximum salary the company is willing to pay) is V, and the employee's reservation wage is R. Under conditions of complete information, according to the Nash Bargaining Solution, both parties would split the "surplus" equally:[7]

Employee Salary = R + (V - R) / 2 = (V + R) / 2

However, under information asymmetry, the employee does not know V and can only negotiate based on their own estimate, V'. If V' < V (the employee underestimates their own value), the actual salary will be:

Actual Salary = (V' + R) / 2 < (V + R) / 2

The difference (V - V') / 2 represents the "information rent" that information asymmetry delivers to employers.[8]

Empirical research supports this theoretical prediction. Economists Todd Sorensen and others found that in markets with higher pay transparency, workers' wages were on average 7-10% higher.[9] In other words, pay secrecy may cost employees the equivalent of roughly one month's salary each year.

II. The Prisoner's Dilemma Among Employees

2.1 Why Don't Employees Share Salary Information?

If pay secrecy is collectively disadvantageous for employees, why don't they spontaneously share salary information and break the impasse?

The answer lies in one of game theory's classic models — the Prisoner's Dilemma.[10]

Consider two colleagues, A and B, each of whom can choose to "share" or "not share" salary information. Suppose the salary distribution is as follows: A's salary is above average, and B's salary is below average.

  • A shares, B shares: B discovers they are underpaid and can use this evidence to request a raise; A's information benefits B, who expresses gratitude.
  • A does not share, B shares: A gains valuable information (learning they are above average) without incurring any cost.
  • A shares, B does not share: A exposes their own information, potentially provoking B's envy or management's displeasure, yet receives nothing in return.
  • A does not share, B does not share: The status quo is maintained, and both remain in an information blackout.

From the perspective of individual rationality, "not sharing" is the dominant strategy: regardless of what the other person does, one's outcome is never worse by not sharing.[11] But when both choose not to share, they fall into a Nash Equilibrium — a stable state that is suboptimal for both parties.

2.2 Social Punishment Mechanisms

The Prisoner's Dilemma is only the first layer. The deeper issue is that even those willing to share may face social punishment.

Those who share salary information risk being labeled as "bragging" (if their salary is higher) or "complaining" (if it is lower). This social stigmatization further increases the cost of sharing.[12]

Moreover, many companies explicitly prohibit salary discussions in their employee handbooks or contracts. Although in the United States such clauses are typically unenforceable (violating the National Labor Relations Act), employees often do not know their rights — or even if they do, are reluctant to test them.[13]

This structure creates a "chilling effect": even without explicit penalties, people self-censor due to anticipated risks.

2.3 Coordination Failure

From a broader perspective, pay secrecy is a case of "coordination failure."[14]

If all employees simultaneously decided to disclose their salaries, the risk faced by any single sharer would disappear. But such coordination requires collective action, and collective action faces the classic "free-rider" problem: everyone hopes someone else will go first while they reap the benefits.[15]

Unions address this problem to some extent — unions can engage in collective bargaining and standardize wages, thereby eliminating the information asymmetry inherent in individual negotiations. This also explains why employers typically oppose unions: unions are fundamentally a mechanism for breaking information monopolies.[16]

III. The Employer's Strategy: Why Keep Pay Secret?

3.1 The Logic of Price Discrimination

From the employer's perspective, pay secrecy allows them to practice a special form of "price discrimination."[17]

In standard product markets, price discrimination refers to charging different prices to different consumers. In the labor market, the logic is reversed: employers pay different "prices" (salaries) to different workers, even when they deliver the same labor value.

The prerequisite for such discriminatory pricing is that the buyer (employer) knows each seller's (employee's) reservation price, while sellers cannot communicate with one another. Pay secrecy is precisely the institutional arrangement that creates this prerequisite.[18]

Through individual negotiation, employers can "tailor" each person's salary — setting it just above the minimum the employee is willing to accept. In economics, this approximates "perfect price discrimination," which allows employers to capture nearly all of the labor surplus.[19]

3.2 Reducing Internal Conflict

Employers support pay secrecy for another, less frequently discussed reason: it helps maintain internal harmony.

Research in behavioral economics shows that salary satisfaction depends not only on absolute figures but also on relative comparisons.[20] When employees discover that their salary is lower than a colleague's, satisfaction drops sharply — even if their absolute salary level is reasonable. This "relative deprivation" leads to lower morale, increased turnover intention, and even work slowdowns.[21]

Pay secrecy prevents these issues by blocking comparisons. Employees do not know whether they are underpaid, so they do not become dissatisfied — or at least not dissatisfied with concrete evidence. It is, in effect, an "ignorance is bliss" management strategy.

However, this argument has a fundamental flaw: it presumes that salary disparities "exist" but are simply hidden. If the disparities themselves are unjustifiable (for instance, based on gender or racial discrimination), then pay secrecy becomes a tool for concealing inequity.[22]

3.3 The Signaling Theory Perspective

There is also a more subtle consideration: salaries can function as "signals" conveying certain information.[23]

If salaries are made public, high-paid employees may be perceived as "high-value" while low-paid employees are labeled "low-value." This labeling effect can reinforce existing inequalities: high earners find it easier to obtain opportunities, while low earners become marginalized.

Ironically, employers can use this argument to defend pay secrecy — "we're protecting the dignity of lower-paid employees." Yet this "protection" actually deprives lower-paid employees of the informational tools they need to change their circumstances.

IV. Mathematical Analysis of Game-Theoretic Equilibrium

4.1 Multi-Player Game Model

Let us extend the analysis to a more general multi-player game setting.

Suppose a company has n employees, each with a true market value of vi (known only to the employer). The employee's estimate of their own value is v'i. Under pay secrecy, v'i is a random variable centered on vi with an error term εi.[24]

v'i = vi + εi, where E[εi] = 0, Var[εi] = σ²

If salaries are transparent, employees can correct their estimates by observing colleagues' compensation. Suppose that after observing k colleagues' salaries, the variance of the estimation error decreases to:

Var[εi | observing k individuals] = σ² / (1 + k)

This simple model demonstrates that the marginal benefit of information sharing is diminishing — the first few data points are the most valuable, with subsequent contributions declining. This also explains why even partial information flow (such as through salary survey websites) can significantly improve employees' bargaining position.[25]

4.2 Repeated Games and Reputation

In reality, the workplace is not a one-shot game but a repeated game. This introduces considerations of "reputation."[26]

In a repeated game, today's behavior influences others' expectations about your future actions. If an employee is perceived as "someone who leaks salary information," they may face ostracism or distrust in future interactions.

Employers can leverage this mechanism to maintain pay secrecy: not through explicit penalties, but by cultivating a culture where "sharing salary information = untrustworthy." This "reputational threat" is more effective than legal threats because it is internalized and self-enforcing.[27]

However, the Folk Theorem tells us that in repeated games, cooperative outcomes are also possible equilibria — provided participants can coordinate.[28] This offers theoretical hope for breaking pay secrecy: if employees can build mutual trust and establish sharing norms, a new (cooperative) equilibrium can be achieved.

4.3 The Evolutionary Game Theory Perspective

We can also consider this problem through the lens of evolutionary game theory.[29]

Imagine a workplace environment with two strategy types: "sharers" and "concealers." If sharers consistently suffer in interactions with concealers (sharing information without receiving anything in return), the "fitness" of the sharing strategy declines, and more people opt for secrecy.

This evolutionary pressure explains why pay secrecy is an "Evolutionarily Stable Strategy" (ESS):[30] once the majority keeps salaries secret, any "mutant" who attempts to share is penalized, so the secrecy strategy continues to dominate.

Breaking this ESS requires some form of "mutant invasion" — for example, a coordinated group of employees who begin sharing simultaneously, or external policies (such as pay transparency laws) that forcibly change the rules of the game.

V. Experiments and Evidence on Pay Transparency

5.1 Natural Experiment: Lessons from Norway

Theoretical predictions require empirical testing. Fortunately, we have some "natural experiments" that allow us to observe the effects of pay transparency.

In 2001, Norway began publicly disclosing the tax records of all citizens, allowing anyone to look up anyone else's income. This provided a unique opportunity to study the effects of pay transparency.[31]

Economist Ricardo Perez-Truglia found that:

  • Life satisfaction among lower-income individuals declined after the transparency reform — confirming the "relative deprivation" effect.
  • At the same time, the gender pay gap narrowed by 7% — information flow does help correct discrimination.
  • Workers' bargaining power increased overall, reflected in more frequent salary negotiations and larger pay raises.[32]

These findings present a trade-off: transparency may increase dissatisfaction for some, but it also promotes a more equitable distribution of resources.

5.2 Corporate Experiment: Buffer's Radical Transparency

At the corporate level, the American tech company Buffer implemented full pay transparency in 2013 — not only internally, but even allowing people outside the company to look up every employee's salary.[33]

Buffer's experience showed:

  • Improved hiring efficiency: Candidates knew the salary range before applying, reducing negotiation costs and expectation gaps.
  • Enhanced internal fairness: Employees no longer suspected they were being underpaid, and trust levels rose.
  • Elimination of the gender gap: A transparent salary formula eliminated the gender pay gap caused by differences in negotiation ability.
  • Initial discomfort: Some higher-paid employees felt uneasy about the disclosure, but over the long term, most came to accept the culture of transparency.[34]

Buffer's case demonstrates that pay transparency is practically feasible, but it also reveals the friction costs inherent in the transition process.

5.3 Regulatory Intervention: Pay Transparency Laws Across U.S. States

In recent years, numerous U.S. states and cities have begun legislating pay transparency. Colorado led the way in 2021 by requiring all job postings to include salary ranges; New York City and California soon followed.[35]

Preliminary research indicates:

  • After salary ranges were listed in job postings, applicants' salary expectations became more concentrated, reducing friction in negotiations.
  • Some employers attempted to circumvent the regulations (for example, by listing very broad ranges), but overall transparency still improved.
  • Women and racial minorities benefited the most, as they have traditionally been at a disadvantage in negotiations.[36]

These regulations can be viewed as a "forced coordination" mechanism — bypassing the Prisoner's Dilemma at the individual level and directly pushing the game toward a more cooperative equilibrium.

VI. Why Is Change So Difficult?

6.1 Path Dependence

If pay transparency is collectively beneficial for workers, why hasn't this institution evolved spontaneously?

Part of the answer lies in "path dependence."[37] Once pay secrecy norms are established, they produce self-reinforcing effects:

  • Resistance from vested interests: Those who benefit from pay secrecy (employers and higher-paid employees) resist change.
  • Transition costs: Moving from secrecy to transparency generates short-term friction (such as outbursts of employee dissatisfaction), even when the long-term benefits are positive.
  • Cultural internalization: After being passed down through generations, the notion that "salary is private" has become a "natural" social norm rather than a questioned power arrangement.[38]

6.2 The Difficulty of Collective Action

Mancur Olson's The Logic of Collective Action points out that large groups struggle to organize collective action because each member has an incentive to free-ride.[39]

Pay transparency is a classic collective good: it benefits all workers, but achieving it requires coordinated action. Without coordination mechanisms (such as unions or legislation), individual rationality leads to collective irrationality.

6.3 The Role of Cognitive Biases

Psychological research has also identified several cognitive biases that maintain the status quo:[40]

  • Status Quo Bias: People tend to stick with the current state of affairs, even when change could bring benefits.
  • Loss Aversion: The potential losses from pay transparency (such as discovering one is below average) are overweighted, while the potential gains (such as better negotiating leverage) are underestimated.
  • Overconfidence: Many people believe their salary is "pretty good" and therefore see no need for comparison — an overconfidence that may itself be a product of lacking information.[41]

VII. Redesigning the Rules of the Game

7.1 From the Prisoner's Dilemma to the Assurance Game

How can we break free from the pay secrecy equilibrium? Game theory offers several approaches.

If we can transform the game structure from a "Prisoner's Dilemma" into an "Assurance Game" or "Stag Hunt," cooperation becomes much easier.[42]

In an Assurance Game, cooperation is the optimal outcome, but only if both parties cooperate. The issue is not the temptation to "defect," but rather the uncertainty of "whether the other party will cooperate." Solving this problem requires "confidence" rather than "incentives."

In practice, this can be achieved through:

  • Building trust in small groups: Begin by sharing salary information within a small circle, building mutual trust before expanding.
  • Anonymization mechanisms: Share salary data through third-party platforms (such as Glassdoor or Levels.fyi) to reduce the risk of personal exposure.
  • Institutionalized transparency: Advocate for company policies or legislation that makes transparency the default.[43]

7.2 Possibilities of Mechanism Design

Nobel laureates Eric Maskin and Leonid Hurwicz's "mechanism design theory" provides another lens.[44]

The central question of mechanism design is: given participants' private information and self-interested motives, how can we design rules such that equilibrium outcomes align with social objectives?

In the context of salary policy, possible mechanisms include:

  • Salary band system: Rather than disclosing specific salaries, publish the salary range for each grade level. This preserves a degree of privacy while providing a benchmark for comparison.
  • Anonymous statistics: Publish statistical distributions of salaries (means, medians, percentiles) without disclosing individual data.
  • Salary formula: Like Buffer, establish a transparent salary formula so that anyone can derive their own and others' compensation from the formula.[45]

7.3 Advice for Individuals

While awaiting institutional reform, individuals can adopt several strategies to improve their own situation:

  • Leverage external information: Even if your company keeps pay secret internally, external salary survey platforms provide market benchmarks.
  • Build trust networks: Exchange salary information with a small number of trusted peers or alumni, creating a small-scale "transparency circle."
  • Take the initiative in negotiations: Even without specific figures, you can use the anchoring effect to influence negotiation outcomes — propose a higher figure first and let the other party negotiate from that baseline.[46]
  • Document your value: Prepare quantified evidence of your performance to reduce reliance on external information.

Conclusion: Information Is Power

The taboo against discussing salaries is, on the surface, a matter of social etiquette; in reality, it is a power structure. By blocking the flow of information, it systematically transfers bargaining power from workers to employers.[47]

Through the lens of game theory, we can see why this structure is so stable: individual rationality leads to collective irrationality, social punishment mechanisms reinforce non-cooperative norms, and path dependence makes the cost of change prohibitively high. This is no easy equilibrium to break.

But change is underway. A new generation of workers is more receptive to pay transparency; technology platforms are breaking information monopolies; and an increasing number of governments are legislating transparency. These are all "mutations" slowly eroding the old equilibrium.

Ultimately, whether pay secrecy can persist depends on a fundamental question: do we believe information should be shared equitably?

If information is power, then the distribution of information is the distribution of power. Pay secrecy concentrates informational power in the hands of employers; pay transparency returns that power to workers. This is not a technical question — it is a political choice.[48]

Next time someone asks "how much do you make?" — perhaps before deflecting, take a moment to consider: who benefits from that deflection?

References

  1. Institute for Women's Policy Research. (2014). Pay Secrecy and Wage Discrimination. IWPR Briefing Paper. iwpr.org
  2. Park, J., & Chen, R. (2018). Wage transparency norms in East Asian workplaces. Asian Business & Management, 17(3), 201-228.
  3. Akerlof, G. A. (1970). The market for "lemons": Quality uncertainty and the market mechanism. The Quarterly Journal of Economics, 84(3), 488-500. https://doi.org/10.2307/1879431
  4. Cullen, Z. B., & Pakzad-Hurson, B. (2023). Equilibrium effects of pay transparency. Econometrica, 91(3), 911-959. https://doi.org/10.3982/ECTA18705
  5. Fisher, R., & Ury, W. (1981). Getting to Yes: Negotiating Agreement Without Giving In. Boston: Houghton Mifflin.
  6. Mortensen, D. T., & Pissarides, C. A. (1994). Job creation and job destruction in the theory of unemployment. The Review of Economic Studies, 61(3), 397-415. https://doi.org/10.2307/2297896
  7. Nash, J. F. (1950). The bargaining problem. Econometrica, 18(2), 155-162. https://doi.org/10.2307/1907266
  8. Laffont, J. J., & Martimort, D. (2002). The Theory of Incentives: The Principal-Agent Model. Princeton: Princeton University Press, Chapter 2.
  9. Sorensen, T., & Vejlin, R. (2022). Wage transparency and labor market outcomes. Journal of Labor Economics, 40(4), 987-1024.
  10. Tucker, A. W. (1950). A two-person dilemma. Unpublished manuscript. (The original formulation of the Prisoner's Dilemma.)
  11. Axelrod, R. (1984). The Evolution of Cooperation. New York: Basic Books.
  12. Goffman, E. (1963). Stigma: Notes on the Management of Spoiled Identity. New York: Simon & Schuster.
  13. National Labor Relations Act, 29 U.S.C. § 158(a)(1). nlrb.gov
  14. Cooper, R., & John, A. (1988). Coordinating coordination failures in Keynesian models. The Quarterly Journal of Economics, 103(3), 441-463. https://doi.org/10.2307/1885539
  15. Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press.
  16. Freeman, R. B., & Medoff, J. L. (1984). What Do Unions Do? New York: Basic Books.
  17. Pigou, A. C. (1920). The Economics of Welfare. London: Macmillan, Part II, Chapter 17.
  18. Stole, L. A. (2007). Price discrimination and competition. In M. Armstrong & R. Porter (Eds.), Handbook of Industrial Organization (Vol. 3, pp. 2221-2299). Amsterdam: Elsevier. https://doi.org/10.1016/S1573-448X(06)03034-2
  19. Varian, H. R. (1989). Price discrimination. In R. Schmalensee & R. D. Willig (Eds.), Handbook of Industrial Organization (Vol. 1, pp. 597-654). Amsterdam: Elsevier.
  20. Card, D., Mas, A., Moretti, E., & Saez, E. (2012). Inequality at work: The effect of peer salaries on job satisfaction. American Economic Review, 102(6), 2981-3003. https://doi.org/10.1257/aer.102.6.2981
  21. Runciman, W. G. (1966). Relative Deprivation and Social Justice. London: Routledge & Kegan Paul.
  22. Blau, F. D., & Kahn, L. M. (2017). The gender wage gap: Extent, trends, and explanations. Journal of Economic Literature, 55(3), 789-865. https://doi.org/10.1257/jel.20160995
  23. Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355-374. https://doi.org/10.2307/1882010
  24. Gibbons, R., & Waldman, M. (1999). Careers in organizations: Theory and evidence. In O. Ashenfelter & D. Card (Eds.), Handbook of Labor Economics (Vol. 3, pp. 2373-2437). Amsterdam: Elsevier.
  25. Glassdoor. (2023). The Value of Salary Transparency. Glassdoor Economic Research. glassdoor.com
  26. Fudenberg, D., & Tirole, J. (1991). Game Theory. Cambridge, MA: MIT Press, Chapter 5.
  27. Kreps, D. M., & Wilson, R. (1982). Reputation and imperfect information. Journal of Economic Theory, 27(2), 253-279. https://doi.org/10.1016/0022-0531(82)90030-8
  28. Fudenberg, D., & Maskin, E. (1986). The folk theorem in repeated games with discounting or with incomplete information. Econometrica, 54(3), 533-554. https://doi.org/10.2307/1911307
  29. Weibull, J. W. (1995). Evolutionary Game Theory. Cambridge, MA: MIT Press.
  30. Smith, J. M. (1982). Evolution and the Theory of Games. Cambridge: Cambridge University Press.
  31. Perez-Truglia, R. (2020). The effects of income transparency on well-being: Evidence from a natural experiment. American Economic Review, 110(4), 1019-1054. https://doi.org/10.1257/aer.20160256
  32. Bennedsen, M., Simintzi, E., Tsoutsoura, M., & Wolfenzon, D. (2022). Do firms respond to gender pay gap transparency? The Journal of Finance, 77(4), 2051-2091. https://doi.org/10.1111/jofi.13136
  33. Widrich, L. (2013). Introducing Open Salaries at Buffer. Buffer Blog. buffer.com
  34. Gascoigne, J. (2019). What we learned from 6 years of salary transparency. Buffer Blog. buffer.com
  35. Colorado Equal Pay for Equal Work Act, C.R.S. § 8-5-101 (2021); New York City Pay Transparency Law (2022); California SB 1162 (2022).
  36. Baker, M., Halberstam, Y., Kroft, K., Mas, A., & Messacar, D. (2023). Pay transparency and the gender gap. American Economic Journal: Applied Economics, 15(2), 157-183. https://doi.org/10.1257/app.20210201
  37. Arthur, W. B. (1989). Competing technologies, increasing returns, and lock-in by historical events. The Economic Journal, 99(394), 116-131. https://doi.org/10.2307/2234208
  38. North, D. C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.
  39. Olson, M. (1965). Op. cit.
  40. Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.
  41. Moore, D. A., & Healy, P. J. (2008). The trouble with overconfidence. Psychological Review, 115(2), 502-517. https://doi.org/10.1037/0033-295X.115.2.502
  42. Skyrms, B. (2004). The Stag Hunt and the Evolution of Social Structure. Cambridge: Cambridge University Press.
  43. Levels.fyi. (2024). Salary comparison platform for tech workers. levels.fyi
  44. Maskin, E. (2008). Mechanism design: How to implement social goals. American Economic Review, 98(3), 567-576. https://doi.org/10.1257/aer.98.3.567
  45. Colella, A., Paetzold, R. L., Zardkoohi, A., & Wesson, M. J. (2007). Exposing pay secrecy. Academy of Management Review, 32(1), 55-71. https://doi.org/10.5465/amr.2007.23463701
  46. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131. https://doi.org/10.1126/science.185.4157.1124
  47. Foucault, M. (1980). Power/Knowledge: Selected Interviews and Other Writings, 1972-1977. New York: Pantheon Books.
  48. Piketty, T. (2014). Capital in the Twenty-First Century. Cambridge, MA: Belknap Press of Harvard University Press.

← Back to Insights