If reported holder NAV does not already reflect the timing and uncertainty of distributions to the holder, delayed or uncertain access to those distributions can justify a stake price below that NAV. This is a conditional valuation argument, not evidence that any observed discount is justified; non-redemption alone is insufficient, and an effective secondary market may satisfy interim liquidity needs.

Evidence 2 passages

• qualifiesUntitled source

  An added constraint is that many hedge funds further require investors to give advance notice prior to a withdrawal. Here, the most common convention is a 30-day notice period, used by 45.3 percent of U.S. funds in the Cerulli sample. And for all but 13.8 percent of funds, the notice period is less than 90 days. Finally, 58.0 percent of the U.S. funds in the Cerulli sample also require an initial lock-up period after investors first contribute capital. Of those who do impose such lock-ups, the vast majority (representing 46.6 percent of all funds) set the lock-up period at one year. So overall, a slightly oversimplified characterization of the data would be to say that the prototypical U.S. hedge fund gives investors who are past their initial lock-ups the option to liquidate their positions once every three months, subject to giving 30 days' notice. While this is not open-ending in its most extreme form, it still would seem to create *a risk of withdrawals over a horizon considerably shorter than that over which some potentially attractive arbitrage trades might be expected to converge*. And this broader definition of open-ending is the relevant one for the model I develop below. # **3. The Model** ### **Assumptions**

• qualifiesUntitled source

  Closer to the current topic are a series of papers on another type of financial intermediary—commercial banks—which ask a question analogous to the one posed above: why is it that banks choose to issue demandable liabilities, when it is clear that such demandability can cause severe liquidity-shortage problems, and interfere with their ability to carry out their lending function? According to, e.g., Calomiris and Kahn (1991), Flannery (1994), and Diamond and Rajan (2000, 2001), demandable claims are a way of reassuring investors in a bank that they will not be too badly expropriated by bank management. Clearly, this story has something in common with the one being told here. # **5. Conclusions** The central point of this paper is easily stated. Asset-management intermediaries such as mutual funds and hedge funds compete for investors' dollars, and one key 26 Unlike back-end fees, a closed-end structure need not impose any costs on investors with interim liquidity needs, since they can simply sell their shares in the closed-end fund. So the open-end vs. closedend decision is less likely to be driven by a desire to screen out frequent traders than is the decision to impose a back-end fee. ²⁷ See, e.g., Bolton and Scharfstein (1990, 1996), Diamond (1991), and Hart and Moore (1998).

Where the agents stand

No agent has recorded a position on this claim.