A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Present bias is the specific manifestation of hyperbolic discounting in which individuals place disproportionate weight on immediate payoffs relative to future payoffs, even when this conflicts with their own long-term preferences. It is captured by the beta parameter in the quasi-hyperbolic model, where beta < 1 means the present is overvalued relative to any future period. Present bias generates two distinct agent types: "naive" agents who do not anticipate their future self-control problems (and therefore fail to take precautions) and "sophisticated" agents who recognize their bias (and may adopt commitment devices or, paradoxically, may give up on long-term goals prematurely). Present bias is the proximate cause of procrastination, undersaving, overconsumption, and many health-related self-control failures.
Present bias is the engine behind many of the self-control failures that define everyday life. It is the reason you hit the snooze button despite planning to wake up early, the reason you scroll your phone instead of starting the report, and the reason you have a second dessert despite committing to a diet. In each case, the immediate payoff (comfort, entertainment, taste) overwhelms the delayed payoff (productivity, career advancement, health) — not because you do not value the future, but because the present is disproportionately weighted in the moment of choice.
The formal model makes the pattern precise. In the beta-delta framework, a person at time 0 evaluates future utility as: U₀ + beta * (delta * U₁ + delta² * U₂ + delta³ * U₃ + ...). When beta = 1, this is standard exponential discounting. When beta < 1, there is a discrete penalty applied to all future utility — as if the present gets a bonus that no future period receives. This means the trade-off between period 0 and period 1 is distorted (the agent requires a larger premium to wait), while the trade-off between period 1 and period 2, viewed from period 0, is not distorted by beta (both are future). But when period 1 arrives and becomes the new "present," the beta discount now applies to period 2, recreating the same distortion. This is the source of time inconsistency.
The naive-versus-sophisticated distinction has rich implications. Naive agents believe their future selves will behave as their current self plans — they are "present-biased optimists." They perpetually intend to start saving, exercising, or studying "next period," creating the classic pattern of good intentions followed by repeated postponement. Sophisticated agents correctly predict their future present bias — they know that their future selves will face the same temptation and succumb similarly. This foresight can lead to demand for commitment devices (valuable self-binding) or to the pessimistic conclusion that long-term goals are unattainable (premature abandonment).
The interaction between present bias and the economic environment determines outcomes. Environments that make immediate gratification easy and delayed gratification hard — abundant fast food, frictionless online shopping, infinite social media feeds — amplify the behavioral consequences of present bias. Environments designed to support long-term goals — automatic savings deductions, friction in spending (e.g., removing one-click purchasing), cooling-off periods for major purchases — can mitigate those consequences. This is the core insight of behavioral public policy: you cannot change beta, but you can change the environment in which beta operates.
Research on present bias has particular relevance for understanding poverty. Mani et al. (2013) showed that poverty itself taxes cognitive bandwidth, and the constant trade-offs required by scarcity may exacerbate present bias by depleting the self-regulatory resources needed to resist immediate temptation. This suggests that the self-control failures commonly observed among low-income populations are at least partly a consequence of their economic circumstances rather than a cause — an insight with important implications for the design of social safety nets and financial assistance programs.