Heterogeneous-agent asset pricing is a class of models in which prices are set by investors who differ in wealth, risk tolerance, beliefs, or constraints, so that the marginal pricer is not a single representative agent. Equilibrium returns reflect the distribution of agents, not just aggregate consumption.
How it works
Standard representative-agent models price assets off one stochastic discount factor tied to aggregate consumption. Heterogeneous-agent models relax this: agents face idiosyncratic income risk, borrowing constraints, or divergent beliefs, so the SDF reflects a wealth-weighted aggregation across the cross-section. The marginal investor — whoever is unconstrained and willing to absorb the next unit of risk — sets the price, and that identity shifts with the wealth distribution.
Why it matters now
As 2025-2026 markets concentrate into mega-cap equities and passive flows, the question of WHO the marginal pricer is — constrained households, price-insensitive index funds, or levered specialists — has moved from academic curiosity to a live driver of risk-premium compression and fragility.
Example
In Constantinides–Duffie (1996), adding uninsurable idiosyncratic income shocks that are countercyclically volatile lets the model generate a large equity premium without an implausibly high risk-aversion coefficient — resolving part of the Mehra–Prescott equity-premium puzzle that the representative-agent benchmark, requiring risk aversion above 30, could not.
Frequently asked
- What is heterogeneous-agent asset pricing?
- Heterogeneous-agent asset pricing is a modeling approach where asset prices emerge from investors who differ in wealth, beliefs, risk tolerance, or constraints, rather than from a single representative agent. The equilibrium stochastic discount factor aggregates across this cross-section, so the marginal pricer's identity — and the wealth distribution — directly shape risk premia.
- How does it differ from representative-agent models?
- Heterogeneous-agent models differ from representative-agent models by allowing investors to face idiosyncratic risk and binding constraints, so no single agent's consumption summarizes the economy. Representative-agent models price assets off one aggregate stochastic discount factor; heterogeneous-agent models require a wealth-weighted aggregation, which can generate larger equity premia and richer cross-sectional return patterns.
- Why does heterogeneous-agent asset pricing matter for markets now?
- Heterogeneous-agent asset pricing matters because identifying the marginal investor — constrained households, price-insensitive passive funds, or levered specialists — explains risk-premium compression and fragility better than aggregate models. In 2025-2026, equity concentration and passive flows make the marginal-pricer question central to understanding mega-cap valuations and correlation dynamics.
- Does heterogeneous-agent modeling solve the equity premium puzzle?
- Heterogeneous-agent modeling partially resolves the equity premium puzzle. Constantinides and Duffie (1996) showed that uninsurable, countercyclically volatile idiosyncratic income shocks can produce a realistic equity premium without implausibly high risk aversion. The representative-agent benchmark of Mehra and Prescott (1985) instead required risk-aversion coefficients above 30, widely viewed as unreasonable.