Kyle's 1985 insider trading model has a single asset, a single informed trader, and a single dimension of private information. Extending it to many assets — each with correlated payoffs, each potentially revealing information about the others — seemed to require many-dimensional complexity. The informed trader's optimization becomes a problem in infinite-dimensional Bayesian game theory, where the strategy space grows with the number of assets.
Jeon, Malamud, and Villeneuve (arXiv:2602.21125) show that the solution collapses to a single scalar fixed point. Regardless of how many assets are traded, how the payoffs correlate, or what aspects of the cross-sectional structure the insider knows, the equilibrium trading strategy, price impact, and informational efficiency are all characterized by one number.
The mechanism: the informed trader's optimal disguise depends on how much information prices reveal, which depends on the trading strategy, which depends on the disguise. This circular dependence, which in finite dimensions would generate a system of coupled equations, reduces in the continuum limit to a single self-consistency condition. The continuum simplifies rather than complicates because asymmetries between individual assets average out.
The result: closed-form characterizations of equilibrium in a setting that appeared intractable. The dimensionality of the problem is not the dimensionality of the solution. Adding assets adds complexity to the setup but not to the answer.
The general observation: in problems with circular strategic dependencies, taking the limit of many interacting components can reduce the solution's dimension rather than increase it. The complexity lives in the specification, not in the equilibrium. When the interactions are symmetric enough, the many-dimensional problem has a one-dimensional answer.