Dynamic Resource Allocation with Hidden Volatility

Felix Zhiyu Feng and Mark M. Westerfield
Journal of Financial Economics 2021, 140(2): 560-581

[PDF] [Journal Version] [SSRN Version]

Summary

We study firms’ internal resource allocation when a manager privately controls volatility and may extract private benefits. The optimal contract is implemented with a constant pricing schedule, and prices are not risk-adjusted. We apply the model to internal capital markets and transfer pricing.

Abstract

We study a firm’s internal resource allocation using a dynamic principal-agent model with endogenous cash flow volatility. The principal supplies the agent with resources for productive use, but the agent has private control over both project volatility and resource intensity and may misallocate resources to obtain private benefits. The optimal contract can yield either overly risky or overly prudent project selection. It can be implemented with a constant pricing schedule (i.e., a static, decentralized, linear mechanism), giving the agent control over the resource quantities, project risk, and agent’s equity share. The implementation rationalizes the use of hurdle rates above a firm’s cost of capital and transfer prices above marginal cost, while showing that hurdle rates or transfer prices may not vary with the agent’s risk choice.

Cite as

Feng, Felix Zhiyu, and Mark M. Westerfield. 2021. “Dynamic Resource Allocation with Hidden Volatility.” Journal of Financial Economics 140(2): 560–581. https://doi.org/10.1016/j.jfineco.2020.12.006

BibTeX

@article{FengWesterfield2021,
  author  = {Feng, Felix Zhiyu and Westerfield, Mark M.},
  title   = {Dynamic Resource Allocation with Hidden Volatility},
  journal = {Journal of Financial Economics},
  year    = {2021},
  volume  = {140},
  number  = {2},
  pages   = {560--581},
  doi     = {10.1016/j.jfineco.2020.12.006},
  url     = {https://doi.org/10.1016/j.jfineco.2020.12.006}
}

The PDF posted here is the authors’ manuscript (February 2020 draft). The version of record is available from the journal at the DOI above. Updated October 5, 2026.