[joschmo]

To play devil's advocate, the reason a private equity firm should use a continuation is that it wants to hold an asset past the fund's 10-year life because the asset is continuing to perform very well. 10 years ago, many private equity firms felt heartburn from letting big software winners go after 5 years because that's the typical fund cycle.

In reality, LPs and GPs would have been better off reliably holding the same growing asset for 15-20 years as the company went from $10M in EBITDA to $250M. Why take a 3x in 5 years when you can get a 25x in 15 years? The annual compounding works out to be roughly the same. The only issue is your fund structure so you trade it amongst yourself every 3-5 years at a 3-4x mark that makes you look good to your investors. The growth is usually there so it is not a lie though liberty may be taken with multiples which expanded rapidly over the last decade.

This phenomenon has been used aggressively, perhaps too aggressively, in software and particularly software roll-ups where you can comfortably compound 20-30% EBITDA growth annually between 10-15% organic growth (half or more of which is just price increases) and an aggressive M&A strategy that drives higher margins.

That said, this article is raising an early and important alarm bell to the broader public, who have a vested interest because their pensions are invested in these firms, that certain firms may be using it too aggressively. When I scan the broader PE landscape, the firm that has used it to the most success is Clearlake Capital. I don't have insight into their whole portfolio but have heard criticisms that they overuse it. Or that their is incestuous trading of assets between them, TA Associates, Hg, Insight Partners, FP, Thoma, and a handful of others.

[SilverBirch]

> The only issue is your fund structure so you trade it amongst yourself every 3-5 years at a 3-4x mark that makes you look good to your investors. The growth is usually there so it is not a lie though liberty may be taken with multiples which expanded rapidly over the last decade.

The problem is that it's not possible to actually say that. Because the valuation never gets tested against the market you're just marking your own homework. It's a strong protection of these closed end funds that at some point you do actually have to provide the return on investment.

To put it another way - who needs to engage in this, well performing funds that are beating their expectations and making big returns, or poor performing funds who need to come up with creative ways to disguise their bad investment decisions.

[ahtihn]

> Why take a 3x in 5 years when you can get a 25x in 15 years?

3x in 5 years is equivalent to 27x in 15 years.

Obviously it's not that simple in practice but it's not like it's easy to predict growth over a 15 year period.

[eftychis]

The issue here is the lack of opportunities. You have a finite number of bins to place your money in. If you find a "winner," there is no necessarily refresh to get new winners.

After some amount invested that likely most people will never see one has a hard time finding suitable places to reinvest -- that is experienced by eg the renessaince fund, Warren buffet etc. Or of course you can inflate the public or private market

[danielmarkbruce]

"roughly" is in the next sentence...and if we are being so pedantic "equivalent" is the wrong phrasing for everyone except like 10 people.

[danuker]

> "equivalent" is the wrong phrasing

A 15-year period is 3 times longer than a 5-year period.

5x3 = 15

Compounding the 3x return, for 3 times:

3^3 = 27

They are exactly, mathematically equivalent.

But of course there's no way to predict that the growth rate will stay the same. Logistic functions look like exponentials, until they suddenly don't (market saturation).

[danielmarkbruce]

The IRR is equivalent. You can't eat(spend) IRR.

The NPV is not equivalent (unless you happen to have a cost of capital = 25%... like about 10 people). The cash on cash is not equivalent. Every person on planet earth will take the 15 year compounding (again except the 10 or so..).... hence they are not equivalent. When returns are high, investors are not indifferent to time horizon. Longer is better. To make it extremely clear - would you prefer IRR of 50% for 1 minute or 10 years?

The only statement which is precise enough is "the IRR is equivalent". Anyone can be pedantic, it rarely helps.

[danuker]

You are right. Only in a zero-interest-rate environment are they "equivalent" WRT NPV. If you can safely get high returns, then you have to discount them.

[danielmarkbruce]

They are equivalent NPV wise only when the cost of capital is 25%. Cost of capital is a fuzzy concept and different folks have different numbers based on all kinds of things (despite the precise numbers put in spreadsheets). From a practical perspective they will almost never be NPV equivalent.

[TrackerFF]

I'd imagine that every acquisition comes with a non-negligible acquisition cost, which should be baked into the calculation.

Less companies held over a longer time would mean less acquisition costs, while more companies held over a shorter time would mean more acquisition costs.

[Bubble_Pop_22]

> When I scan the broader PE landscape

How do you do that? What's the source for getting to know what's going on with PE firms and their funds?

[1jbdg]

IRR is higher with 3x over 5 years than 25x over 15 years. More fees in the first example