Incubator funds

Incubator funds (also known as incubated funds and limited distribution funds) are created by mutual fund companies through a trial process typically seeded with the companies' own capital or employee capital. Incubator funds are privately operated and are not open to the public, although they are registered with the SEC. Registration insures that, if the incubated fund is eventually opened to the public, performance data can be reported and advertised. Fund companies only open the top performing funds to the public. The higher performing funds that survive the incubation period are used by the fund company to generate business. The funds with unattractive performance, which would be more difficult to market, are liquidated.

Research studies on incubator funds
Evans (2009) examines fund data from 1996 - 2005 and finds that for 1,048 newly created funds, 242, or approximately 23%, were incubated. During the incubation period, Evans found that these funds outperformed non-incubated funds by 9.8% annually. When adjusted for risk, the outperformance ranged from 1.4 percent to 3.5 percent, with a more than doubled Sharpe ratio. However, three years after opening to the public, incubated funds underperformed non-incubated funds by 3.2 percent annually, and had lower Sharpe ratios. Additional findings in the study include the following:


 * Incubated funds have higher net flows than non-incubated funds, and that the higher flows are consistent with investors following historical outperformance.
 * Broker distributed mutual funds (load funds) are more likely to incubate new funds, especially in an investment objective where the fund family's current offerings are attracting lower net dollar flows.
 * The addition of incubator funds into the fund universe introduces a bias in reported fund returns. Over the 1996 - 2005 period Evans found a statistically significant bias for four-factor returns of 0.43% annually; a statistically significant bias of 0.84% annually for equal weighted returns; and a statistically insignificant bias of 0.11% for value weighted returns. Evans argues that applying a ticker-creation date or age filter removes the incubation returns bias.
 * An ostensible justification for incubating funds is an attempt by fund families to identify superior managers and superior investment strategies. This justification is belied by the finding that manager tenure is shorter in families that incubate and by the subsequent underperformance of incubated funds once they go public.
 * Looking at the year by year results, the percentage of incubated observations and the impact of incubation on aggregate fund performance are both larger in the early part of the sample. Possible explanations for this decrease over time are (1): backfilling (Evans extends the data series to include 2006 and 2007 and does not find this to be a sufficient explanation) and (2): a movement away from using incubation as a performance based strategy for fund family asset gathering. Other means of competing for the investor investment dollar include fund proliferation (ICI data from 2005 show that between 1996- 2004 the total number of mutual funds increased from 6,248 to 8,044, an increase of 28.7%), and fee differentiation (the same ICI data show mutual fund share classes increased 93.5% during the 1996 - 2004 period, from 10,352 to 20,036).

Garavito (2008) examines a number of characteristics of incubator fund management and incubator fund asset holdings. The author finds incubated funds hold more illiquid and less-popular securities and are better at buying stocks in the incubation period than after their Initial Public Offer (IPO). Incubated funds also take more concentrated bets and invest in stocks that are not part of the typical institutional investor stock universe. During the incubation period, incubated funds can pursue such strategies since there are no investors who can withdraw funds on demand. Once funds go public, they must keep a certain level of liquidity to meet potential withdrawals.

Palmiter and Taha (2009) point out the following preferential treatment fund companies provide incubator funds in the quest to produce superior returns.
 * The fund companies will overallocate initial IPO stock offerings to the incubator funds. Because the incubator funds are small, even a small allocation to hot IPO's can significantly increase returns.
 * An incubator fund does not have to worry about fund redemption. It does not need to hold cash reserves. It can thus hold riskier, illiquid, thinly traded stocks. The funds' small size can magnify the fund manager's winning security selections (made by either skill or luck).
 * The fund companies kill off underperforming incubator funds; the winning funds are opened to the public.