Standard economic theory holds that money is fungible — a dollar earned from freelance work should be treated identically to a dollar from a salary. In practice, decades of behavioral economics research suggest otherwise. People routinely sort money into mental categories based on where it came from, and those categories shape how it gets spent.
Thaler's Framework
The concept of mental accounting was formalized by economist Richard Thaler in a 1985 paper published in Marketing Science, and expanded in his 1999 review in the Journal of Behavioral Decision Making. Thaler proposed that individuals maintain implicit mental accounts — separating, for instance, "regular income" from "windfall" — and apply different spending rules to each. Money labeled as a bonus or unexpected gain tends to be spent more freely than money perceived as earned through labor.
Thaler's framework drew on earlier prospect theory work by Daniel Kahneman and Amos Tversky, particularly their observation that people evaluate gains and losses relative to reference points rather than in absolute terms. Mental accounting extends this by showing that reference points shift depending on the label attached to the money, not just the amount.
Fungibility Violations in Practice
Empirical research has confirmed that fungibility violations are widespread. Economists Justine Hastings and Jesse Shapiro published a 2013 study in the Quarterly Journal of Economics examining how households respond to gasoline price changes. Their data showed that consumers treat savings from lower gas prices differently than equivalent salary increases — spending the "freed up" gas money on premium fuel rather than redirecting it to general household needs.
A separate line of research by Abigail Sussman and Rourke O'Brien, published in the Journal of Marketing Research in 2016, found that consumers categorize debt differently depending on whether it was incurred for necessities or discretionary purchases. The labels attached to the underlying transaction — not the interest rate or balance size — influenced repayment prioritization.
In the context of supplementary income, these findings are relevant because money earned from a secondary source is often psychologically separated from primary earnings. Research by Hal Hershfield and colleagues at UCLA has explored how temporal framing — thinking of future income as present or distant — affects saving rates, though this work focuses on retirement decisions rather than supplementary earners specifically.
The Labeling Mechanism
More recent experimental work has tried to isolate the labeling mechanism. A 2019 study by Anja Achtziger, Carlos Alós-Ferrer, and Alexander Wagner, published in the Journal of Economic Behavior & Organization, used laboratory experiments to test whether simply changing the label on an identical monetary amount altered spending behavior. Participants who were told they had "earned a bonus" spent a larger fraction of it on hedonic goods compared to participants told the same amount was "part of their payment."
The effect was not universal. Participants with higher financial literacy scores showed smaller labeling effects, suggesting that awareness of mental accounting may partially counteract it. However, the study sample was drawn from German university students — a demographic that limits generalizability to broader populations or to people earning supplementary income in non-laboratory settings.
Methodological Limitations
Several well-documented concerns apply to this body of research. First, most laboratory studies on mental accounting use relatively small stakes — often under fifty dollars — and it remains unclear whether the same effects hold when the amounts involved are large enough to materially affect household budgets. Field studies like Hastings and Shapiro's partially address this, but those designs introduce confounding variables that are difficult to control.
Second, much of the foundational research was conducted with samples from Western, educated, industrialized, rich, and democratic (WEIRD) populations, as noted by Joseph Henrich and colleagues in their widely cited 2010 paper in Behavioral and Brain Sciences. Whether mental accounting operates similarly in non-WEIRD contexts — particularly in economies where supplementary income is the norm rather than the exception — is an open question.
Third, the line between mental accounting and rational behavior is not always clear. Earmarking money for specific purposes can function as a self-control device, as Thaler himself acknowledged. In that framing, treating bonus income differently from salary is not a bias but a deliberate strategy. The empirical challenge is distinguishing between the two explanations in observational data.
What the Research Does and Does Not Show
The evidence suggests that the source label attached to income influences how households allocate it. This pattern has been replicated across multiple research teams and study designs. What the research does not establish is whether this behavior produces net positive or negative financial outcomes — that depends on individual circumstances and cannot be generalized from laboratory or survey data.
Steady Ledger does not draw prescriptive conclusions from these findings. The purpose of this report is to describe what researchers have observed and where the evidence is incomplete.