Shared investment

A fund joins several things: capital committed together, authority delegated to a manager and exposure to a common portfolio. Those functions are useful, but they answer different problems.

The Formation Game asks which functions an investment requires and what it costs to separate them. It studies investments that need several participants, portfolios with different permissions and the consequences of deciding early or keeping a right to refuse.

Principal findings

Capital compatibility

Capital must be permitted to finance the investments that need it. In the paper’s allocation model, feasibility depends on whether every subset of investments is covered by enough compatible capital. A failed test identifies the specific bottleneck.

Allocation flexibility

In the paper’s large-population models, independent differences in preferred exposure can offset one another. The flexible capital needed grows with the square root of the population under the stated assumptions. A common component of demand can instead require a fixed share of capital.

Commitment timing

A late right to refuse can turn small differences in fit into a common funding shortfall. Earlier binding participation and appropriately designed transfers change that exposure. Replacement capital, cash for mechanism payments and flexible investment authority perform different jobs.

Verifiable information

The lower bounds for efficient, truthful allocation with a retained veto depend on favored projects being private. When those projects are externally verifiable, a procurement auction can pay investors to move, with no withdrawals and bounded expected cash payments in the specified benchmark. Those payments still require funded cash. Recording a preference does not make it verifiable.

The costs of separate exposure

Shared records can reduce the compensation needed to induce investigation under the paper’s incentive assumptions. Administration and equal-treatment clauses also affect the cost of individual terms. Conventional funds can implement relevant arrangements and belong in the comparison.

Assumptions and scope

The paper contains several models. Its conclusions depend on the information, contracts, preferences and funding available in each. The two-project benchmark is not a general claim about every private investment.

The results separate investment funding, assurance against formation failure, discretion over allocation and replacement capacity. Numerical checks examine the specified models; they do not measure investment performance.

Collective investigation appears in the paper’s delegation analysis. Endogenous search over new opportunities and supplier terms remains a further research question. The paper does not establish universal superiority over conventional funds or guarantee access to a particular asset.

Publication details

The Formation Game. John Komkov. October 2026. This download preserves the manuscript supplied for the 8 October 2026 website snapshot.

The complete manuscript contains the definitions, proofs, appendices and references. The source and replication package accompany this edition.

John Komkov has commercial interests in Argosy and Megalith. This website’s synopsis and research organization were prepared with AI assistance.

Earlier research and supporting material

References

See the manuscript for the full bibliography. Related reading includes What belongs in a fund? and the separate AI methods report.