Private Equity and Venture Capital Fund Performance: Evidence from a Large Sample of Israeli Limited Partners

Pension funds and institutional investors around the world have been allocating an increasing fraction of their assets under management to private equity (PE), venture capital (VC), and other types of private funds. Public pension funds tracked by Preqin, for example, have steadily increased their allocations to this asset class over the past decade, with the median allocation rising from 18.1% in 2010 to 30.3% in 2020, and 79% of investors stating that they expect to allocate a larger proportion of their funds to private equity by 2025.

In our paper, Private Equity and Venture Capital Fund Performance: Evidence from a Large Sample of Israeli Limited Partners, publicly available on SSRN, we study the performance of non-US based limited partners. The performance of non-US limited partners may differ from what has been documented for US limited partners for various reasons, such as differences in access to top performing funds, differences in skill or ability to select successful PE or VC fund general partners, and differences in fees.

Our study relies on a unique and comprehensive data set of on all capital calls and distributions associated with investments in PE, VC, and other types of private funds for a 20-year period ending in December 2019 by the eight largest institutional investors in Israel, managing 76% of all retirement savings. This cash flow data constitutes part of the information that the institutions managing retirement savings are mandated to report to the Capital Market, Insurance, and Savings Authority at the Ministry of Finance. The data set is therefore free of survivorship and other biases documented in the literature in the context of some commercial data sources on PE and VC fund performance. Much like their peers elsewhere, institutional investors in Israel, including pension funds, life insurance plans, and other forms of long-term savings known as provident funds, have increased their allocations to illiquid assets from 12% of assets under management in 2010 to 17% in 2020. Their investments in PE, VC, and other types of private funds have increased from a mere 1% of their assets under management in 2010 to 5% in 2020. This increase in the allocation of funds to PE and other funds coincided with a dramatic increase of 250% in the total value of assets under management during this time period, driven by the introduction of mandatory retirement savings. As a result, the volume of investment by Israeli LPs in PE and other types of private funds has become economically large in absolute, not only relative, terms.

The performance of Israeli pension funds is interesting primarily because it is likely to be indicative of the performance of non-US based institutional investors elsewhere. The pension system in Israel is comprehensive, where every employee and employer must deposit a fixed fraction of the employee’s income each month. These long-term savings are managed by for-profit, non-bank institutions, primarily insurance companies and other non-bank investment managers. In many ways, this pension system resembles the retirement savings plans of many developed (OECD) countries. In addition to its structure, the Israeli pension system is interesting because of its rapid growth. The main consequence of the introduction of a mandatory retirement saving system in Israel has been an increase in AUM of close to 10% p.a. over the past decade. The Israeli pension system may offer lessons for similarly structured, fast growing, retirement savings systems in other countries.

We find that that PE performance in our data set is slightly lower than what has been documented for US limited partners in the literature. It is also lower than the average performance of funds listed in Preqin and Burgiss in our sample time period. One possible interpretation for the lower performance is limited access: Israeli LPs may not be able to gain access to the top performing (“top quartile”) funds, possibly because Israeli limited partners are, on average, smaller than US limited partners. Limited access may also be driven by the Israeli limited partners’ remote geographic location relative to the majority of PE funds and perhaps by the perception of some fund managers that limited partners based outside North America are less prestigious than leading local (US-based) LPs. Prior investment experience may also affect access: established LPs may be better able to form connections with established fund managers. Israeli LPs are relatively new players in the PE market, although we do not find much evidence to support the conjecture that their access improves over time. Conceptually distinct from limited access, limited partners’ performance may also depend on skill in identifying and selecting top performing first-time funds and general partners. We provide evidence that is consistent with both access and skill playing a role in explaining the inferior performance of Israeli LPs relative to the stylized facts on the performance of US-based LPs. We do not find evidence that these limited partners pay higher fees relative to their US-based peers.

We next compare the performance of investments in local, (within-Israel) versus foreign (in the US and elsewhere) funds. Hochberg and Rauh (2013) argue that, in the context of US states, political interference and home bias in investments result in poor performance of local, within-state, investments of US public pension funds in PE relative to their out-of-state investments. By contrast, Morkoetter and Schori (2021) gather international data and find that IRRs in foreign PE and VC funds, defined as funds located outside the limited partners’ home region, are lower than the realized IRRs in local, within-region, funds. Our evidence lends support to the latter view: we find that local, within-Israel, PE and other private funds consistently outperform foreign funds. This result holds across virtually all fund types (the only exception is hedge funds) and is especially pronounced for PE funds, real estate, and infrastructure funds. We attribute this finding to the limited partners’ superior access to top performing local funds, as well as to their skill in selecting local first-time funds, which appears to be better than their ability to select foreign funds.

Finally, we also examine the possibility that the difference in returns realized by limited partners in our data set and those reported by US-focused studies may be related to the nature of coverage in Preqin (we also make some comparisons with Burgiss), databases commonly used in the literature. Although it has been argued that data derived from Preqin and Burgiss are generally not severely biased, we find a large discrepancy in performance between the universe of funds that are held by limited partners in our sample and the subset of these funds which are also included in Preqin. Funds that are missing in Preqin are not necessarily non-US based but are typically small funds that exhibit poor performance relative to funds which are included in Preqin. This suggests that, while Preqin may be useful in estimating the returns generated by the PE/VC industry as a whole, it may not reflect precisely the returns realized by limited partners who choose to, or are forced to, invest in relatively small funds that are unlikely to be top performers.

The full paper is available for download here.