Recycling Riches: The Devil is in the Detail
Should GPs recycle? Is recycling good for LPs? Like everything in venture capital and private equity – it depends.
As always, this is not investment advice. Do your own due diligence and form your opinions.
A few years ago, I co-invested alongside one of the most highly regarded VCs in Silicon Valley. It was an AI company. The product was probably ahead of its time, and so product-market fit was elusive. It became clear it would not become the 5x, 10x+ outcome we and our partner VC had underwrote. It was marked at around cost. Understandably, the founders were also starting to think about their next steps, after years of a startup salary. In the end, we voted to exit to a strategic acquirer for about that cost. We got our money back - actually, a bit less.
It made sense to move on. Traction was there. Founders were fatigued. The market changed quicker than we expected. But I couldn’t quite square it. Why not just keep having “free” shots on goal? What’s a couple million on a $200m fund?
It was only later that the final piece of the puzzle clicked: recycling.
Why do it?
Simply put, when a portfolio company is realized, rather than distributing those proceeds back to LPs, it is “recycled” in the fund by using it investing in new portfolio companies, or meeting the follow-on reserves of existing ones.
Many venture and private equity funds recycle , or at least have a provision in the governing documents that allows it to recycle capital. Usually, its around 20-30%.
Usually, recycling is not controversial (but it can be - more on that later). It can be good for both the LP and GP. This is usually a non-issue, and some of the most aggressive recyclers are some of the best and most prudent VCs.
This has three main benefits.
It allows the entire fund size to be put to work (and sometimes more). A $100m fund will have, say $15m, in fees over a 10-year life. That leaves $85m in investible capital. By recycling, the GP can invest the full $100m. An LP that commits $100m would have the full $100m put to work.
The LP makes fewer re-investment decisions (usually a good thing). There’s conversation that gets around HNW and family office investors new to venture and private equity – liquidity. VC and PE is illiquid by nature and the excess return the access class can generate is inherent to its liquidity – it must be illiquid to generate those returns.
What fewer LPs at the end of that market contemplate, is that more liquidity often means less opportunity to compound capital at high rates of return for longer. Not always but often. If you do all the work, build the relationships, to finally allocate to (and be accepted by) a top tier VC, I’d rather let them compound returns than try to find another fund that meets a high hurdle rate.
Of course, there are exceptions to the rule and I’ll go into those later.
It optimizes for absolute net return on LP dollars (“you can’t eat IRR”). Generally, and particularly in venture, the goal is a maximize your net multiple of investment over a reasonable period. IRR is important (despite my earlier point) , the key is a reasonable amount of time, not an indefinite amount of time) but absolute dollar-on-dollar return is paramount.
Example Recycling Route
Let’s illustrate number 3 above. This will make it clear how recycling actually works. We’ll assume this GP achieves 3x of return for every dollar it invests. If it’s a $100m, $85m is investible (as per the example above). With recycling, it can invest the full $100m.
We see that by recycling, the GP has amplified the net returns by putting more capital to work at 3.0x gross return. Of course, this of course assumes the GP can find 3.0x returns with that recycled capital. Put another way, the GP would need only 2.5x gross return with recycling to achieve net 2.2x, rather than 3.0x gross return to achieve the same net 2.2x.
However, this extends the distributions into the future, which may lower IRR depending on the timing of cash flows, the amount of recycling, and the actual gross return.
I’ve modeled a very simple cash flow schedule using these assumptions. In this example, if we split incremental increase in future distributions (260 - 244 = 36) over two final distributions.
If we increase the recycled capital from 15 to 40, we get an even higher net TVPI of 3.2x and the same IRR as 31%.
So why not recycle into perpetuity?
Firstly the example above is only mathematical. In reality:
Recycling capital does not necessarily mean distributions are moved a couple years into a the future. It might be many years later and LPs now earn something like 3.2x over 15 years.
There may be no distributions at all on that recycled capital. Or lower than 3x gross. The maths above makes a big assumption: the GP can keep earning 3x gross on every incremental dollar invested. This is not the case - and a good GP will size the fund (incl. recycling) appropriate to the available investment opportunities that generate the target return. At some point, incremental capital invested will generate lower returns.
LPs may want distributions. They have may have found opportunities with an even higher hurdle rate than venture (tell me where those are) or need to balance their portfolios (too much PE, too little of other).
Recycling Red Flags
Now let’s consider another recycling parable. This time not so good.
I recently screened a PPM for the second fund of a new VC. Four things stood out immediately, before I can even got to the investment strategy:
Fund I had recycled almost 50% of its committed capital;
That fund was small, relative to the cheque sizes required for its strategy;
The recycled exits were between 2-3x exits each; and
the GP’s carry hurdle was based on a TVPI multiple rather than IRR.
Taken together, these are red flags.
Flag 1
I suspect the fundraise didn’t go as well as planned, and the fund was too small for the stated strategy. The first fund is critical (there’s an argument that says all successful VCs today are just survivors of a good first vintage). The GP needs a solid track record, with respectable amounts of capital invested, to raise the next fund.
That’s fine, but make sure your LPs are happy with more exposure than planned. Make sure your fund II investors don’t get turned off, or at least understand your rationale for fund I.
Flag 2
Recycling is best used to return the capital from investments that have remained at, or around cost (without further upside), to redeploy into potential future home runs.
Harvesting 2x or 3x returners is risky, as it may cap upside unless: (a) it is very clear that such upside is unlikely; and (b) the new invested (recycled) capital outperforms.
The GP needs to be very confident that (a) and (b) above are true, and the gross performance had not yet warranted high conviction on (b).
You want another shot on goal for a 10 bagger, not the harvest of average returns.
Flag 3
Show me the incentive and I’ll show you the outcome. The carry hurdle on a multiple of net capital is not bad in itself. In fact, I like that. I’m optimizing for cash-on-cash returns, over a reasonable period, when allocating to a VC.
However, on a fund that is not going as well as hoped, I do not want the GP recycling ad nusuem to hit a carry hurdle.
I’ll show you why, again with some easy math. It will also nicely put the three red flags into perspective.
Small fund, say $40m of commitments, with $6m of fees, leaving $34m of invested capital. Performance hasn’t gone as planned, so let’s say every dollar of invested capital is a 2x gross return. The GP earns carry over a 1.2x net TVPI.
The net TVPI is higher than the gross! How can the net number, after fees and carry, be lower than the gross number? Because the numerator of both net TVPI and gross MOIC has increased (because both include realized and unrealised value), but only the denominator of gross MOIC has increased (because only MOIC uses invested capital, TVPI is paid in capital from LPs).
So the GP has earned $11m in carry, rather than $4m, without better raw investment performance (gross MOIC is 2x in both cases).
Put another way, do you want your GP to re-invest capital at the 2x times it seems to be earning, or do you want the distribution to allocate to another manager?
What if this other manager recycles only 20% of its capital, and achieves the same net TVPI of 2.32x? Same net return I know, but I’d argue that’s a better manager. It would have a better gross MOIC, a measure of raw investing skill, which is more likely to have persistence for future performance.
Finito - Summary for LPs
So GPs are wise to recycle up to say, 30% of the fund size, provided the available investment opportunities can generate sufficient (the target) gross return. It is therefore best used for the realisations of returns that are marked at, or around 1x, and it is very clear that the upside beyond that is limited, to be recycled into potentially outperforming investments.
It is best used in venture capital, when the GP has raised a disciplined fund size and it is used to put the full fund size to work (and maybe more than that). Many prudent GPs recycle the management fees, essentially offsetting them so the invested capital is near the fund size.
LPs should check the fund documents for limits and review how recycling has been used in prior funds.
If it is used excessively, perhaps to meet a shortfall in the expected fund size, in conjunction GP incentives and less than stellar gross returns, LPs should make sure they understand the GP rationale then and in future.