CALLING CAPITAL

CHIEF FINDINGS

  • PE funds tend to call only about ⅓ of committed capital during the first 2 years of a fund’s life, which creates a need for sophisticated liquidity planning
  • There is substantial variability in the pace at which PE funds call capital
  • Call pacing appears to strongly vary by fund vintage, with moderate cyclicality
  • At a generic level, buyout and VC funds have practically identical call pacing
  • At more granular levels, there are noticeable differences in call pacing between the two

COMMITMENTS & LIQUIDITY

Private Equity (PE) is generally considered to be an illiquid asset class: once investors transfer capital to a PE fund, they typically cannot access it again until it is distributed by the fund manager – usually towards the end of the fund’s planned life (which is typically around 10 years, but can regularly be extended). We won’t delve into all the minutiae of obligations and interactions between PE fund managers (‘general partners’, or GPs) and their investors (‘limited partners’, or LPs). Our focus in this post is on the dynamics of how LPs transfer capital to GPs.

At the beginning of a PE fund’s life, LPs commit to investing pre-specified amounts of capital. But they don’t transfer the entirety of those commitments to GPs immediately. Instead, GPs ‘call’ commitments in increments over the fund’s life – mostly in its early years as they’re investing in assets. LPs thus face a liquidity management problem: once a LP makes a capital commitment to a PE fund, it’s contractually/legally obligated to transfer portions of that commitment when asked to do so by the fund’s GP (i.e., whenever a call is made). The LP could choose to be fully conservative, and keep its commitment available as cash. In doing so, however, it would miss the opportunity to earn higher returns on the amount of the commitment by investing in non-cash assets. This opportunity cost to conservatism is known as ‘cash drag’. An alternative would be to invest portions of the commitment in assets that have higher upside potential than cash, while still preserving liquidity (e.g., public equities). But the challenge here is that higher returns are very hard to align with liquidity uncertainty: by investing committed capital in such assets, an LP takes on some risk of being unable to fulfill a call from its GP.

This liquidity problem is mitigated when LPs can accurately forecast the timing and size of capital calls that come from their GPs. When armed with accurate call predictions, LPs can make more efficient (and less risky) use of committed capital before it is called – thus increasing their overall performance. How accurate can such forecasts be? And what might the gains be to LPs in making use of them? To begin answering these questions, we will first characterize the time profiles of capital calls, and how they vary across different PE funds.

OBSERVATIONS

We start our exploration by looking at PE capital calls ‘at large’ – i.e., not restricted to fine-grained time periods, geographies, or fund sizes. This will give an overall benchmark against which we can compare the call dynamics of more specific fund types. The dataset which we use is comprised of the annual call profiles (that is, what percentage of committed capital is called in each year of a fund’s life) of more than 2000 buyout and venture capital (VC) funds that were launched from 1998-2018 (the year in which a PE fund is launched is commonly referred to as its ‘vintage’). We restrict our analysis here to buyout and VC funds because their portfolios are each comprised of the same type of asset: private companies. This makes for more direct comparisons across funds than would be the case if we were to include other types of private market funds, such as timber, real estate, or infrastructure. (We plan to explore those other branches of private capital later, in another post.)

FIGURE 1: AVERAGE CALL PACING: ALL FUNDS
(1998-2018)

The above chart is a so-called ‘box-and-whiskers’ plot: boxes for each fund year reflect the inter-quartile range (IQR) of call sizes for that year – i.e., the top of a box marks the cutoff for the largest 25% of calls, and its bottom marks that for the smallest 25%. The white horizontal bars in between those cutoffs are the average call sizes for their corresponding years; the horizontal white bars outside of those cutoffs are the top and bottom 10% markers (respectively) for the given year.

Two features of the chart are immediately salient. First, over the past two decades, buyout and VC funds have called about 1/3 of the capital committed to them within the first 2 years of their lives, and 1/2 by year 3. This fact evidences the need for LPs to undertake sophisticated liquidity planning, to ensure year 4 and beyond are managed correctly. LPs might be able to justify keeping commitments only in cash if GPs called nearly all of that capital in the first 12-24 months of a fund’s life. Yet, in reality, the bulk of commitments are called after the first 24 months, which suggests that the effects of cash drag may be significantly depressing LPs’ investment returns.

A second salient feature in the above chart is that, over the first 5 years of funds’ lives, there is substantial variability in the pace at which funds call capital. Do these features hold for VC and buyout funds separately?

FIGURE 2: AVERAGE CALL PACING: VC FUNDS
(1998-2018)
FIGURE 3: AVERAGE CALL PACING: BUYOUT FUNDS
(1998-2018)

The above figures demonstrate that, across vintages for 1998-2018, the call profiles for generic (i.e., not segmented by size, geography, etc.) VC and buyout funds are indeed highly similar. It bears reiterating that this similarity is at the population level – that is, averaged over all VC and buyout funds (respectively) in the X-SPEEDS dataset. Consequently, from the standpoint of planning for capital calls, there is little categorical difference between investing in VC or buyout funds, especially if one can diversify across vintages (note: the results above are averaged across funds that launched between 1998 and 2018, inclusive).

However, when one zooms in to more granular levels, noticeable patterns of difference emerge. Among the more subtle of these are distinctions across vintages, as depicted below.

FIGURE 4: AVERAGE VC CALL PACING BY VINTAGE
(YEARS 1-5 OF FUND LIFE)
FIGURE 5: AVERAGE BUYOUT CALL PACING BY VINTAGE
(YEARS 1-5 OF FUND LIFE)

It’s noticeable from the above chart that call pacing can swing dramatically with fund vintage (especially in the first 4 years of a fund’s life), irrespective of whether it’s a VC or buyout fund. These swings partly map to macroeconomic cycles. Specifically, notice the speedups in calls that precede both the Dot-Com Bust of 2000 and the Global Financial Crisis (GFC) of 2007-2008, as well as the slowdowns that follow those events. As might be expected, in the lead-up to the Dot-Com Bust, acceleration in calls for VC funds exceeds that for buyout funds; meanwhile, the speedup for buyouts is more dramatic prior to the GFC. This difference is likely due to the different roles each type of fund played in those two events. Notably, the acceleration among buyout funds’ calls prior to the GFC could be ascribed to the full deal pipelines that were touted by fund managers from 2004-2007). Likewise, one can spot a less pronounced acceleration in buyout funds’ calls 1-2 years after the Dot-Com Bust – when companies damaged by the Bust became relatively cheap (and therefore enticing for buyout investors).

Further, post-2000, buyout funds have, on average, been less consistent in how they call capital in years 2-4 than VC funds.

Differences in how buyout and VC funds call capital are also partly driven by the geographies that they target. As the figures below show, buyout funds that focus on Asia and Europe tend to call capital at a faster rate than their VC counterparts (there is no statistical difference between buyout and VC funds that concentrate on US-based investments).

FIGURE 6: AVERAGE VC CALL PACING BY GEOGRAPHY
(2009-2018; YEARS 1-5 OF FUND LIFE)
FIGURE 7: AVERAGE BUYOUT CALL PACING BY GEOGRAPHY
(2009-2018; YEARS 1-5 OF FUND LIFE)

Surprisingly, differences in fund size do little to explain differences in call-pacing schedules for either buyout or VC funds (measured in $mm USD in the figures below). That is, call pacing appears to be relatively consistent across fund sizes.

FIGURE 8: AVERAGE VC CALL PACING BY FUND SIZE
(2009-2018; YEARS 1-5 OF FUND LIFE)
FIGURE 9: AVERAGE BUYOUT CALL PACING BY FUND SIZE
(2009-2018; YEARS 1-5 OF FUND LIFE)

WHAT’S AHEAD

In future posts, we will be exploring cash-flow behavior at the end of the fund lifecycle: that is, the timing and scales at which PE funds distribute their capital.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s