Private Markets, Public Opinions

Much has been published recently regarding public versus private valuations. On one side of the isle, many public equity portfolio managers have argued that private valuations are everything from inaccurate to fraudulent.

On the other hand, most professionals engaged in private equity – as users, vendors, practitioners, clients, and providers alike – seem to think the process works just fine as is.

While admittedly imperfect, I think the average PE valuation is fair, and if anything, probably understates the true value of the company across most of the economic cycle. That’s not to say there aren’t areas of overvaluation, or even egregious overvaluation from time to time. We’ll get to this shortly.

PE is also admittedly cyclical and tied to broader economic activity, but even at extremes I believe it more accurately reflects underlying corporate value than public stocks, which are often vastly more over- and under-valued than PE ever is. If something never gets as massively overvalued, then logically it doesn’t need to correct as much on the downside either. Again, we’ll get to that in more detail, too.

I’ll sum up the “volatility argument” from public stock investors as “our lines are squiggly and yours are straight, so yours are wrong.” However, rather than continue the debate, I want to present a few first principles and follow this up by providing data on how PE valuation works in practice. This data has informed my view; take it for what you will.

To begin with, I’d like to point out that:

1)?????Private equity is not one thing

2)?????Company life cycle and market dynamics drive valuations

3)?????The public stock market ≠?the economy

These three thoughts will inform the discussion that follows, and we’ll hit each of these with data, but first let’s look at how PE valuations work in the real world.?

Understanding Private Equity Quarterly Valuations

?

The valuation practice across the industry is not arbitrary, as some assure us it is. All institutional private equity firms have well-established valuation practices. While some pay it lip service, many have formal valuation committees that meet quarterly and review valuation packets in great detail for each business. These processes are often overseen by a fund accountant, controller, or CFO internally, almost always with accounting credentials and many times a CPA. And most of them take their job seriously.

Sometimes, as is common in venture capital, another round or outside investor establishes an explicit valuation for the company based on the price they paid for their minority interest – (this is almost exclusive to venture capital and VC isn’t the private equity industry as a whole. Private equity isn’t one thing!) Other times, GPs will use indications of interest from a banker to establish valuations for a company they are shopping for a sale, generally only when a sale is imminent.

But most of the time, particularly within buyout strategies, there’s a valuation method applied like a Discounted Cash Flow methodology, or a revenue or EBITDA multiple. The process involves applying the multiple to the relevant financial metric and arriving at the new valuation. Now, the multiple was not arbitrarily selected; it was heavily influenced by what the initial multiple was (in many cases, that multiple was held constant for a year or more), as well as comparable recent transactions in the same sector and capitalization range.?

For example, if the trailing twelve-month EBITDA for a small service business was $10 million, and the multiple is 10x, the valuation would then be $100 million. If this is the average multiple the most relevant 15 similar companies traded at over the last 6 years, then it is certainly appropriate regardless of the degree of temporal variability in Tesla and Google’s public valuations.

And at the end of the year, these valuations are approved by an external auditor such as KMPG or Ernst and Young, (who by the way, much prefer the multiple approach to the DCF model, since multiples are what the businesses will actually transact on.) The auditors are imperfect, just as they are in the public company auditing business, but they also take the job seriously, and do push back on valuations that they find egregious.

Some firms include their Limited Partner Advisory Committee in the conversation, sometimes even requiring them to vote on quarterly valuations. I’ve sat on these LPACs before, reviewed quarterly materials in a meeting (including financials and comps), and voted for valuations. (I’ve voted against plenty of other consent items, FYI, but valuations were never even a top 10 concern.)

Anecdotally, in the years I sat on LPACs which had valuation oversight, most of the year-end disputes with auditors involved positions being held below where the auditor thought they should be marked, not vice versa. (In fact, academic research – see here – and practitioner evidence alike – see here ?– suggest this isn’t anecdotal at all. Aside from much of the momentum that is VC, private equity NAV is estimated to be undervalued in aggregate by 20% to 30%.)?

It's worth noting this twice – the best actual empirical evidence on private equity invested capital valuations suggests it’s undervalued.

So, at the fund level, we have companies, entry multiples and entry EBITDA, Trailing Twelve Month EBITDA and current multiples, and hence, current valuation. Now, in most funds, almost every quarter there is a company or two that appears richly valued – perhaps extremely so – but there are reasons, and there is often also a company or two that appear dramatically undervalued. But on average, most everything in the fund is well justified.?

It looks a little like this:

In this case, the fund has a MOIC of 1.8x since inception. While the average multiple has risen from 9.3x to 9.5x, the bulk of the value creation came from EBITDA increasing. On average, EBITDA was up 65% over the hold across the portfolio companies.

Let’s take a look at one specific company, in this case Company 5, which has been marked up 4.4 times, as an example. The fund invested $36.2 million into this business in October of 2020. The total enterprise value was $61.5 million, meaning debt was $25.3 million, and EBITDA was $6.1 million. Fast forward to now, EBITDA has exploded to $24.3 million, up four times – extreme growth, but not terribly uncommon for the big winner in a small fund.

At the same time, the multiple has been marked up from 10.1x to 11.2x. So, multiple expansion has driven the mark-up, but far more of the equity value creation has come from EBITDA growth. In fact, we can quantify this explicitly.

Three turns of value creation came from earnings growth, 0.4 times from the multiple increase, and for the sake of simplicity, we are holding the debt-to-equity constant at 0.588 (which actually means they had to put more dollars of debt on the business and no equity value creation came from debt paydown). This all adds up to the 4.4 multiple of invested capital, which – if funded all at once on 10/24/2020 – yields an eye-popping IRR of 97.7%.

Now, this is not to say these results are typical, but they provide the basis for how to think about valuations and value creation. Which leads us back to the multiple increase. Recall the business was bought on a 10.1x EBITDA multiple in 2020, and it’s been marked up to 11.2x today.?

Usually, there are a series of comparables that determine what multiple should be used. For example, Company 5 has 12 comps in the same sector and roughly similar capitalization above that have actually transacted in private markets, and these comps yield an average valuation that is typically used to both justify the price paid at acquisition – buyers and sellers both use similar comp sets but of course tilt the playing field towards their respective arguments – as well as validate interim valuations over the hold.

Several things should be immediately apparent. Yes, the multiples paid in the sector have risen over time, but not the stratospheric increase often seen in public market P/E ratios during periods of euphoria. The average multiple over the entire period is 10.7x, but since 2019 it has risen to 11.8 times. Additionally, one could argue that the current valuation applied to Company A is indeed conservative, since it is lower than five out of the last six actual market transactions.

Further, when revenue growth rate is considered, one could argue company A might be undervalued relative to the comparable set. Businesses which grow earnings faster year-over-year than their peers, like Company A, should and do trade at higher multiples. In the software space – where companies tend to trade on a revenue multiple as opposed to an earnings one – this is very apparent. The chart below from SaaS Capital plots actual funding transactions for a basket of private SaaS businesses relative to their TTM revenue growth. Faster growing businesses do trade at a premium.?

However, as mentioned, there are often other factors that are less visible than growth to the LPs which the manager is aware of that impact the multiple. It could be revenue quality, it could be profit margins, it could be the management team, and it could even be a fantastic and loyal client base. Better businesses should command higher multiples.

Moreover, we note that the multiple is applied to the TTM financial metric, which itself is autocorrelated and has a smoothing effect on valuations. Hence the valuation increase is far more smoothed on the way up than public equities, which necessarily means it should also be moderated on the way down. When – or more accurately if – earnings decline, the markdown will be driven primarily by the EBITDA changes. Of course, catastrophic earnings declines warrant more significant mark downs, but by definition those are idiosyncratic events.?

Regardless, the point is that these valuations are mostly reasonable and done in good faith. Outside of late-stage venture or expensive software, the multiple is rarely the driver of the valuation increase, and what multiple is the most appropriate is that number which is likely to be the exit multiple when the company will be sold at some indeterminate point over the next three to five years.

Current valuations in public equities are generally not relevant and are excluded from the comparables set when valuing a business. Indeed, there are other structural issues in private equity that are far more important to a company’s valuation than public stock valuations.

Which brings us to the next point…

Private Equity is Not One Thing!!

?

Currently, there are about 4,000 publicly traded companies in the US. According to the US Census Bureau, there are 7.7 million US companies (C-corps, S-Corps, LLCs, and LPs) that have at least one employee. If we constrain this to businesses with $5 million or more in topline, there are perhaps 500,000 companies. However, if we look at those above $10 million topline, it drops to 200,000 and maybe those above $50 million in revenue, there’s only about 25,000. There are a lot of privately owned businesses in the United States.

But how many of these are PE backed? Combining the databases of Burgiss, Pitchbook, and Preqin, there are about 15,000 to 16,000 PE-backed companies currently in the US. Private equity remains a target rich environment and is nowhere near saturated. There are always plenty of new companies to look at for potential investments.

According to data from PitchBook, from 2008 until 2022, there were 65,378 PE acquisition totaling roughly $7.9 trillion worth of capital. Each year, there were around 4,700 deals done, and $562 billion worth of transaction volume for an average deal size of $120 million.?

However, averages in a highly skewed distribution are seriously misleading. Let’s look under the hood at the actual deal flow. 86% of those dollars went to deals of $100 million and up, and a full 43% of the total dollars were in transaction sizes of $500 million plus. By dollars, large transactions dominate private equity.?

However, by number of deals, small transactions lead private equity transaction volume. 45% of total deal count were transactions less than $25 million and 96% were below half a billion. These are very, very different market segments, and they look no more alike than does public and private equity in general.?

Let’s put this back into context – while the data sets don’t perfectly match up, they do coalesce around a picture emerging in private markets. Private equity isn’t one thing; in fact, it looks more like a stratification instead of merely being a binary separation between public and private market segments.

At the small end of the scale, there are tons of businesses, and by deal count PE is most active here. However, it is a lightly fished pond, where less than half a percent of all the potential businesses have private equity backing. As one moves up the capitalization spectrum, the opportunity set gets slightly smaller, PE activity becomes a larger component of capital formation, but the number of deals as a percentage of total PE deals becomes smaller and smaller.

Company life cycle and market dynamics drive valuations

Clearly, it should also be apparent that competitive market dynamics also become different going up market. Private equity market participants are becoming less selective, often trading the same business over and over. Simultaneously, pricing is also very different going from one market segment to the next: small market businesses trade at lower valuations, lower middle market prices are higher, upper middle market even higher and large market typically trades at the highest multiples.?

And this isn’t merely anecdotal; consultant StepStone aggregated entry level multiples across PE dating back to 2000 into two buckets – large market funds and small market funds. This analysis included underlying data from thousands of funds over nearly two decades. (I believe for this piece of research the classification was large funds or those above $5 billion, middle market funds between $2.5 and $5 billion, and small market funds below $2.5 billion. In the interest of simplicity, I omitted the middle category, which had multiples falling right in the middle of the other two. It is worth noting again that there are many more layers to this capitalization spectrum than big/small).

The structural upward creep in the industry is immediately apparent; multiples in both the small and large market have increased at 2.5% to 2.6% per year. No question this has driven PE returns. However, the other apparent phenomenon is that large transactions trade at a premium to small deals every single year.

On average, the larger deals trade at a 20% premium to the smaller transactions, at least 1x more. The only notable exceptions are 2001 and 2009, immediately after recessionary periods, when large funds traded just 0.4 times north of the small funds. Though still a premium, this is not as wide as normal.

The reason why small market multiples fell less dramatically through the Dot.com bubble and GFC is largely because they used less leverage. The same StepStone research decomposed the average multiple paid into the debt and equity components. We can see that large funds used more leverage than small funds in the runup to the GFC, and when credit markets closed, it resulted in more multiple compression for the more levered funds.?

The implication – consistent with our view of market strata – is that larger funds are more cyclical and more tied to capital markets conditions, whereas smaller funds are more insulated from it. This also appears in the fund returns. Using fund data from Burgess, I was able to show that larger funds did mark down more and smaller funds held up better through the cycle.

However, there are other observations to highlight. First, since private equity holding periods are 4 to 7 years, there are very few periods of multiple contraction over any likely hold period, and even less for the smaller funds where multiples are more stable. Private equity allows the option to sell if prices are favorable, but to continue to hold if not.

The other observation is that small companies can grow to become large ones. Another way of putting it is that multiple arbitrage for the majority of these deals in our tables above will not come from public equity valuations but from a larger private equity sponsor buying the company at exit, probably in a new valuation bracket. To wit, even if I bought a business for 7.7x in the small market in 2007 – peak valuations – it’s possible the appropriate exit multiple for my business in 2011 isn’t 7.4x in the small market but 8.7x as I’ve grown up market.

The Public Stock Market Is Not the Economy?

At the risk of sounding like a broken record, let’s turn to some more data. Using data from PitchBook once again from the period 2008 to 2022, we can observe a cumulative 16,432 PE exits totaling nearly $5 trillion in liquidity. The dollar volume is split somewhat evenly between the three different off-ramps: $2 trillion in strategic corporate acquisitions, $1.9 trillion in financial sponsor led deals, and $1 trillion in IPO activity.?

However, once again, by deal count, the picture looks very different. Although 20% of the dollars went to IPOs, just 5% of the deals by count exited in a public listing. The IPO market isn’t relevant for 95% of private equity. Outside of mega deals and $2 billion series E rounds, public valuations don’t really matter because most private companies have at least 3 or 4 more owners and 10 to 15 more years before they could consider a public equity listing if they even wanted to.?

?????????

Today, companies stay private far longer, in no small part because capital formation in private markets is far deeper than it used to be. The last few boom years notwithstanding, the average number of IPOs in the United States has dropped by 80% over the last two decades. And the average age of a company going public today is more like 12 to 14 years old as opposed to 6 to 8 years old in the 90s.

And because of this, private equity owners are able to eat more of the company’s growth cycle. (In fact, I agree with some academics that argue private equity has eaten the public stock market small cap effect.)

If we compare the contribution of earnings growth to equity returns from private equity to public equity over time, this becomes apparent. Using data from Professor Robert Shiller from 1994 to 2021, public stocks have generated an average rate of return around 10.2%. The dividend yield was 2.7% and price appreciation was 7.5%. Additively, that price appreciation came from 1.9% of multiple accretion and 5.6% earnings growth.

Not bad, but not as good as PE. Using proprietary data from multiple sources across thousands of private equity backed companies with the return bridge analysis we introduced above, I think a similar decomposition of PE returns is informative.?

Over the same time period, private equity has averaged 13.6% per year. Interestingly, the leverage effect has generated about 2.7% of the total return, similar to the dividend yield from public equities. However, multiple expansion contributed 4.7% and earnings growth was 6.2%. More multiple arbitrage than public equities, to be sure. However, given the capitalization strata effects discussed earlier, multiple accretion is more predictable when it’s based on company life cycle and predictable structural market pricing than purely market sentiment, dynamics I refer to as manufactured and transitional alpha .?

The private equity industry focus on manufacturing alpha, or predictably and consistently growing revenue and earnings over time, has gradually come to supplant the reliance on leverage or multiple arbitrage as the primary driver of returns.

Research from Ernst and Young going back to the 80s has documented the same phenomena, namely that earnings growth has become a larger and larger component of private equity returns decade over decade. In the 1980s, leverage drove 50% of the total nominal return of PE deals, but in the 2010s that number had dropped to just 12%. At the same time, operational metric improvement has soared from 18% return contribution to 58% last decade.

Anecdotally, I know most PE firms would admit they have to focus more and more on this component of returns as the more reliable and controllable lever for their funds going forward. As one GP puts it “I have to get better faster than the market gets more competitive.

Returning to academic research for confirmation – because I don’t love relying on surveys or anecdotes, in case you couldn't tell – a recent study shows solid support for private equity ownership indeed leading to improved revenue and earnings growth. A paper published in December of 2022 by a pair of scholars from the United Kingdom looked at 1,200 realized buyouts in the UK, and their work was able to show that during the private equity ownership period, these companies demonstrated superior firm sales, earnings, market share and productivity growth relative to a group of control firms. The companies did 30% to 50% better.

In fact, the effect persists beyond the initial control period – potentially due to operational improvements the PE owners made during their hold – and these gains were significant in the smaller and younger portfolio companies. Interestingly, incidence of financial distress via formal insolvency found that PE owned companies were also more likely to experience bankruptcy proceedings after the holding period relative to a control group, but this risk was concentrated in older and larger firms.

(There’s that whole life cycle thing again!)?

You see, in private markets, ultimate value crystallization is all about realized cash flow, not multiples so much. And interim valuations – absent venture capital, where additional funding and subsequent rounds are part of the model – aren’t as informative as financial and operating performance itself. If I know earnings are up 20%, I don’t have to sell for 4 years, and larger market businesses in my sector are still trading at higher multiples than I bought the business for, why would taking those earnings numbers and ascribing some arbitrary multiple +/- 20% tell me more than I already know?

Some GPs will ascribe unrealistic valuations, and some LPs will not have as much transparency as others, but the model is working just fine for the majority of deals. According to the American Investment Council , 86% of all businesses receiving private equity investment in 2020 employed 500 or fewer employees, and more than half had less than 50 workers. For most companies, it’s not about arbing some imaginary public/private discount; it’s about operating the business well and selling at some point down the road to another corporation or another financial sponsor.

And private equity companies empirically grow revenues and earnings faster than other businesses, and empirically tend to be held below fair value. And 95%+ of these businesses are not going to exit public capital markets.

Obviously, there is economic sensitivity in private equity businesses; there is beta to GDP, if you will. And since there is beta to GDP, there is beta to public equities. However, GDP is the third factor variable; public equities of course have beta to the GDP as well. But it is underlying corporate performance – revenue growth/decline, margin increase/decrease, and cash flows that drive both.

Public equity admittedly does a far better job of pricing market sentiment, of pricing aggregate market participant risk appetite in real time, than private equity does, but that’s not necessarily the same thing as the true underlying economic value.

At the highs on February 7, 2023, Google had a market capitalization of $1.39 trillion. Two days later, at the low end of the range, the company had a $1.19 trillion market cap, a loss of $200 billion worth of corporate enterprise value in less 2 days.

Did the fundamental underlying value of Google actually decline by this amount in two days, on no real news? No, but the price did. Such price discontinuities are tangible proof that the public market is imperfect – the price as objective arbiter of real value was wrong either before or after, and possibly both.

Private equity doesn’t need to figure out a clearing price when it’s not transacting. The assumptions that you need to mark the book where it would price if you had to sell it tomorrow is an invalid assumption, because the vast majority of the companies don’t need to sell tomorrow. And any logically correct argument based on an invalid assumption is itself invalid.

Private equity valuations may not be perfect, but they work just fine the way they are. That’s why they won’t change much; because they don’t need to. They aren’t make believe, they aren’t fraudulent. By and large, they accurately represent the economic value of patient capital in private businesses where time is on our side.

I concede there will be more mark downs to the most over-priced segments of private markets, and there will be even more if the recent banking / credit crisis spreads. But PE has the benefit of waiting until it actually happens, and more time to work through it if and when they have to. If they need to be marked down, they will be.

To quote Benjamin Graham one more time, “Market quotations are there for convenience, either to be taken advantage of or to be ignored.”