Please see the preceding posts in our CalPERS Debunks Private Equity series:
• Executive Summary
• Investors Like CalPERS Rely on ILPA to Advance Their Cause, When It is Owned by Private Equity General Partners
• Harvard Professor Josh Lerner Gave Weak and Internally Contradictory Plug for Private Equity at CalPERS Workshop
• CalPERS Used Sleight of Hand, Accounting Tricks, to Make False “There is No Alternative” Claim for Private Equity
• How CalPERS Lies to Itself and Others to Justify Investing in Private Equity
One of the surprisingly widespread behaviors we’ve observed in the limited partner community is the degree to which they choose to believe things that that great social commentator Will Rogers called “just ain’t so.” So CalPERS, in its November workshop to its board members, serves as a lens into the dubious justifications for investing in private equity.
As we indicated, the workshop offered so many targets, in terms of questionable logic and practices, that we’ve struggled to figure out which ones to highlight. Today we’ll review three: the way investors defend and even prefer to rely on flattering general partner accounting over economic reality; the sloppy thinking behind the idea that private equity is a “long term investment”; and yet another eye-popping defense of the flawed return measurement, internal rate of return (“IRR”).
Why Investors Love Phony PE Valuations
Unlike most other types of investment, the unrealized investment in a private equity fund, meaning the companies it has bought and will eventually sell, are priced only by the general partner. There is no third party valuation. By contrast, hedge funds get monthly valuations from independent valuation firms. So the general partners get do what no Wall Street firm would allow to occur on its trading desk, which is have traders to mark their own positions.
The private equity industry does have an argument for this troubling “trust me” setup: that the cost of valuing portfolio companies by an independent party like Houlihan Lokey would be on the order of $30,000 per company. Multiply that by, say, 20 companies four times a year, and you’ve got $2.4 million in costs. On a $1 billion fund, that’s close to a 0.25% annual cost on the commitment amount, which would represent a meaningful drag on returns.
But here’s the flip side: If you allegedly can’t afford to have adequate investor protection, should you be investing in that strategy at all, particularly in light of the considerable latitude that general partners have in computing these estimates?
The dirty secret of these valuations is that jiggering key assumptions, like the discount rate, margins, revenue growth, reinvestment requirements, within ranges that are plausible, results in a very large range of possible values. It’s not uncommon to find that changing the assumptions would result in a valuation of ten times what you’d get using conservative assumptions.
The truth is the portfolio company valuations are nothing more than a general partner’s opinion.
And there is strong evidence that valuations are often phony, as in artificially high, in at least three circumstances:
When stock markets are in distress, as in during the financial crisis. Leverage is a double-edged sword. Borrowing amplifies gains and losses. But you curiously don’t see that in how private equity general partners value portfolio companies, When you see downdrafts in the stock market, which typically are the result of trouble in the real economy that hurts both private and public companies, you don’t see private equity valuations falling as much (on the whole) as stock prices. Instead, the long-established practice in private equity is that when stocks take a nosedive, general partners mark down the valuations of portfolio companies much less. That makes no sense, since in general, private equity firms are levered, and levered equity should fall further, not less, in price than stocks as a whole.*
Around the time of raising a new private equity fund. In a 2013 paper, How Fair are the Valuations of Private Equity Funds? Tim Jekinson, Miguel Sousa, and Rüdiger Stucke concluded that general partners goose their valuations around the time they are raising a new fund, based on data from the entire life of CalPERS’ investments. Those false higher prices make returns look better than they really are:
We find that valuations, and reported returns, are inflated during fundraising, with a gradual reversal once the follow-on fund has been closed. Third, we find that the performance figures reported by funds during fund-raising have little power to predict ultimate returns. This is especially true when performance is measured by IRR. Using public market equivalent measures increases predictability significantly. Our results show that investors should be extremely wary of basing investment decisions on the returns — especially IRRs — of the current fund.
Note that the abstract of the paper also confirms (but underplays) the first point we mentioned, that of understating price declines in weak equity markets, which they benignly call “smoothing.”
Late in the life of the fund. Despite its widespread use, IRR, as we’ve discussed at some length in previous posts, is a poor metric. It also happens to be particularly flattering to private equity. Moreover, its quirks lead investors to have even stronger incentives than they’d ordinarily have to monetize successful investments early in the life of the fund.
As a result, the companies that are left in a fund later in its life tend to be dogs. A common pattern is to report prices for these businesses at their historical valuation level, even though the general partner would have presumably have sold the company by then at that price if that really was the price (the returns would be better if, for instance, a company was sold for $400 million in year 6 rather than in year 9). So these companies are eventually sold, more often than not at a price lower than what they were carried at, with the general partner making plausible-sounding explanations as to why it made sense to sell at a loss versus the valuation.
None of this would be that bad if investors were duly skeptical and took steps to manage against these deficiencies. For instance, they could insist on much stronger rights to demand portfolio company valuations, and have five or ten percent of the companies valued every year, with the selection made on a random basis. Similarly, the Jenkinson paper points out early returns on funds, when measured in IRR, don’t do a good job of predicting eventual performance. Yet despite experts having recommended the use of public market equivalent as clearly superior to IRR, limited partners still prefer to use IRR.
So why don’t investors look at the valuations more skeptically? They view those phony smoothed valuations as a feature, not a bug. Private equity researcher Peter Morris called it “A triumph of accounting form over economic substance.”
One big reason is that when most financial markets are tanking, private equity looks less bad by virtue of under-reporing its price declines.
We discussed earlier in the week, for instance, that in a presentation in the private equity workshop, CalPERS swapped in a phony “observed volatility,” based on those airbrushed valuations, for the better approximation of “expected volatility” that their own consultants use, as well as top players like Goldman, JP Morgan, and Yale. But the use of those “smoothed” values as the basis for those “observed volatility” calculations makes private equity look much less risky than it is. It also creates the incorrect impression that private equity prices move less in synch with the stock market than they really do. (Studies based on actual cash flows from private equity find a 90% correlation, which is considered to be high).
In some ways, I hate singling out this workshop panelist, Bob Maynard, since he’s candid about the issue of the smoothing being questionable. But his remarks illustrate the point: that the investors know full well that this supposed advantage amounts to gaming of reporting systems and the larger political processes that are tied to them, and they embrace this gimmickry.
Allan Emkin, Pension Consulting Alliance: The next question is what do you individually and your boards see as the role of private equity in the portfolio?
Bob Maynard, Chief Investment Officer, Public Employees Retirement System of Idaho: We’re I think more skeptical of private equity than many and actually I’ve been quite surprised at the experience we’ve had, which has been dead solid on the average. Our time-weighted returns are almost exactly yours [turning to CalPERS’ head of private equity Réal Desrochers], that 1.34 above what you could put in the public markets is exactly our experience, so we’ve actually gotten average institutional experience, so it’s worked out better than we were expecting.
We knew we were entering an area where we would not have much influence over what we could do.
Ah, in fact, ah, we recognized however that we were going to get some pressure to look at local investments in private side, we did know that our actuaries and accountants would accept the smoothing that the accounting would do. It may be phony happiness, but we just want to think we are happy and they actually do have consequences for ah, ah, actual contribution rates we are going to be able to put in place. So we’re looking for it even if it just gave public market returns, we’d be in favor of it because it has some smoothing effects on both reported and actual risks, as seen that way.
Once you get into the area, this is kind of a like a rental car return type investing. Once your front wheels are over the spikes, you can’t back up. You’ve got to keep kind of going forward. If we’re going to have a little bit, we’ve got to at least have enough to have a difference in the portfolio, which means get to at least 5 to 10%. And by the time we got to the 2000s, we had gotten to that point, so it did make an appropriate difference. So we’re there, it’s demonstrated benefits in the portfolio, we’re happy if it gives public market returns, anything extra, because of its effect having some smoothing of the risk as seen by the accountants and actuaries and, um, just don’t, you’ve got to keep going once you are there.
As you can see, this is an unusually straight-forward acknowledgment that what is driving the selection of private equity is accounting treatment that Maynard admits is “phony.” And he does not expect to get any other benefit from private equity.
In fairness, the other limited partner on this panel had the opposite view. Chris Ailman, the chief investment officer of CalSTRS, said his institution looked to private equity for returns, and their models depicted it as highly correlated with public equity. That is tantamount to saying that they don’t treat private equity as having lower risk than stocks, so they don’t (or at least don’t profess) to seeing the fudging of the valuations in bad markets as a plus.
But even though Ailman, like the more savvy investors who correctly see that private equity is highly correlated with stocks, and hence offers virtually nil in the way of diversification benefits, CalPERS wants to have it both ways. As we showed in our post earlier this week, CalPERS snuck in “observed volatility” to measure risk, which is based on the accounting values that Maynard correctly depicts as phony. That leads to private equity looking like it is both lower risk, and less correlated with stock market performance, than it really is. Either of those mistaken beliefs would lead you to allocate more of your portfolio to private equity than you would if you used more accurate proxies.
Similarly, seeing private equity returns as rosier than they are, as CalPERS does, will similarly lead to an overly high allocation. As we’ve stressed, and even academics who were formerly private equity stalwarts now admit, the case for private equity returns is not compelling. And there’s a glaringly obvious reason why. This comes from the public comments section, from an MBA student:
Pamela Morris: I had a couple of questions for the board I was hoping you could clarify based on the presentation earlier. In particular, slide 40, I don’t know if we can pull that up at all. It was in Dr. Lerner’s presentation.
So I’m curious based on that slide if the return environment for private equity is in relative and absolute terms as favorable as or similar to or different from it was in the 1990s? I think what we see on slide 40, it seems to suggest that, you have perhaps a problem of too much or more money chasing fewer deals compared to in the past.
If you look at the slide, which is displayed clearly in the video (good for CalPERS’ staff for displaying it), it states “Private equity share of global equity market has grown from 2.7% in 2005 to 5.6% in 2014.” When CalPERS talks about private equity to the media, it makes a point of stressing its returns over the last 20 years. But as slide 40 shows, the world of private equity has changed radically in the last decade.
And in that decade, as we’ve pointed out, CalPERS, as well as CalSTRS, has failed to meet their benchmarks. They’ve also fallen short for every sub-period within 10 years. Failing to meet the benchmark means that by its own measures, these institutions are admitting they do not get paid enough in the way of returns for the risks they are taking.
Yet despite private equity having these clear shortcomings, we see investors again defending private equity based on its fake lower risk/smoothing attributes and the misconstruction of its returns (failing to acknowledge that private equity hasn’t generated enough in the way of returns to justify investing it, per the actual experience of large and sophisticated investors).
In keeping, notice this surprisingly lukewarm endorsement from another academic who has published regularly on private returns in a Los Angeles Times story last week:
Steven N. Kaplan, a finance professor at the University of Chicago, cautioned that although CalPERS’ private equity portfolio has indeed beaten alternatives in the past, even after fees, studies show the rate of outperformance has slowed more recently.
“You should watch it very carefully going forward,” he said.
The “Private Equity is a Long-Term Investment” Myth
We didn’t address this issue yesterday when we took issue with many of the arguments CalPERS listed as showing that general partners had the incentives similar to those of limited partners. That was in part because the issue goes well beyond the fuzzy concept of “alignment of interests”.
Here are some of the key misconceptions:
Pension plans have vastly longer time horizons than private equity. Even though the expected life of a private equity fund is around ten years, that does not mean all the funds committed are at work for ten years. As Oxford professor Ludovic Phalippou, who has studied the actual cash flows in and out of private equity funds, stresses, the money goes in during the first five years close to equally, as in on average 20% per year, and starts being returned in year four to five, again in a gradual manner. He finds that the “average investment life” is close to a mere four years.
By contrast, public pension funds receive payments from the entities that employ workers over their twenty to forty-five year working lives, and then make payouts over their retirement, which extends from five to say, twenty years. You can see these time frames are not all that well matched.
The uncertainty of private equity cash flows are a negative for long-term investors that are managing against actuarial timetables, like pension funds and life insurers. In theory, the ideal investment for life insurers and pension funds is bonds, since you can find truly long-term bonds (20 to 30 years in maturity), and can match the cash flows to a fair degree against expected future payouts. But bonds have defects too, the biggest being the risk of loss of value in an inflationary environment and/or relative to inflation (we have negative real interest rates in a low inflation setting right now), and that argues for having some holdings in investments with higher return prospects (portfolio theory separately argues for diversification by asset class, not to increase returns, but diversification allows you to lower risk).
The high uncertainty of timing of of private equity cash flows is a further complicating factor for long-term investors. You will see that CalPERS regularly point out that a negative of private equity is that they cannot match the actual amount of money outstanding in private equity to the their portfolio allocation. That is due to the fact that when CalPERS has to turn over funds to the general partners (the capital calls) and the distributions are all under the general partners’ control (in finance-speak, CalPERS has given them options). Moreover, when CalPERS gets distributions, and its allocation says the money it got back needs to go back into private equity, CalPERS cannot do that quickly or readily. It takes time to screen and negotiate new fund commitments.
Private equity investors do not take a long-term perspective in managing their investments. One argument you will sometimes hear investors like CalPERS make in favor of private equity is that private equity investors take a longer-term view than public companies, which manage to quarterly earnings. By implication, private equity is more “patient” and therefore more virtuous.
In fact, private equity is often short-termist, but in a different manner than quarterly-earnings-fixated public companies. First, general partners are under pressure to have some early winners, both to boost reported IRRs and to show some actual verifiable profits before they raise their next fund, typically four to five years after their last fund. And remember, that “last fund” is not fully invested day one, but has to compete for and close deals over time.
The result is that private equity firms have engaged in some very short-term strategies, such as dividend recaps, where the general partner would load up a newly acquired company with borrowed funds, and strip the cash out via a giant dividend. The result was often a great initial return, with high odds of an eventual bankruptcy.
Similarly, as much as managing to the quarter is painful, when the time has come to sell a private equity owned company, the point of view is even more compressed. Rather than managing the business to look good at the end of the quarter, the business is being operated to look good every day.
Duff & Phelps’s Barmy Defense of the Use of Internal Rate of Return to Measure Private Equity
On the one hand, CalPERS is to be commended for introducing the concept of public market equivalent, which is a much better valuation method, in this presentation. However, the presentation was too often silent on what approach it was using to measure returns (for instance, in the first section, where CalPERS had a series of charts that compared private equity returns to those of other asset classes).
And while staff might be loath to diss IRR too much, since CalPERS has since 2002 published the returns of all of its funds quarterly, using IRR, it should have at least clued the board in that IRR was problematic and public market equivalent was a much better measure. Instead, this was at the very beginning of the Duff & Phelps presentation:
I was gobsmacked to read the claim, “IRRs are fact based” as was pretty much every finance literate person that I had look at that slide.
Surely David Larsen of Duff & Phelps knows, or ought to know, that at any point in time, the computation of the IRRs in most of the funds in CalPERS’ portfolio have a substantial portion of value left in unsold companies. As we stressed above, those valuations are general partner opinions. Thus the net asset value, which can be the largest item in the IRR computation, is not a “fact.” That means a key input for any fund with meaningful remaining value in the fund is subjective, not objective. The result is that the IRR is indeed “judgment based.”
And this problem is widely recognized, Professor Josh Lerner, who was CalPERS’ star witness at this panel, has said,
When you look at how people report performance, there’s often a lot of gaming involved in how they report the IRR
Similarly, as David Larsen knows or ought to know, the very computation of the IRR involves an embedded assumption. As McKinsey wrote in 2004:
For decades, finance textbooks and academics have warned that typical IRR calculations build in reinvestment assumptions that make bad projects look better and good ones look great….
We believe that managers must either avoid using IRR entirely or at least make adjustments for the measure’s most dangerous assumption: that interim cash flows will be reinvested at the same high rates of return.
So IRR is not unbiased, contrary to his suggestion. To reiterate: that is why academics and savvy investors like the Kauffman Foundation have urged limited partners to use public market equivalent as the best approach for comparing private equity to public investments.
CalPERS’ private equity workshop was a major opportunity lost. While the board members no doubt did pick up some new and valid information, too much of this exercise was designed to validate the status quo. It would have been much better for CalPERS, and the limited partner community as a whole, if this workshop had instead served as a first step in examining the real merits and costs of participating in private equity, and what CalPERS could do to mitigate the shortcomings.
* We pointed out if you had used the same valuation methods for private equity firms as you do for liquid assets, the value of many portfolio companies and the funds themselves would have been zero on September 30, 2008, because there would have been no bids. That matters in the context of using private equity valuations in optimization models, since the data needs to collected on the same basis for the model’s results to be valid. In general, one of the problems with private valuations is the wide-spread practice of using only single-point valuations, as opposed to showing the range of valuations that one would get using different methods and different assumptions.
Really illuminating. As someone who knows quite a bit about the downstream effects of private equity but not enough about the mechanics, these posts are super-helpful.
‘CalPERS snuck in “observed volatility” to measure risk.’
Given the absence of any quantification of volatility in some previous consultant presentations, it would be interesting to know where CalPERS’ 19% observed volatility came from. Specifically, is 19% observed volatility based on a generic index, or is it derived from CalPERS’ actual private equity portfolio?
Even if it’s the latter, capital calls and distributions introduce complexity into volatility calculations that are brute simple for publicly traded shares. Given a series of NAVs, the industry standard approach is to use a chained Modified Dietz formula to account for the varying amount of capital invested in each period.
In a footnote on page 16 of this document, the European Venture Capital Association observes dryly that:
“Unsmoothing techniques,” indeed! As Mae West quipped, “Fasten your seatbelts, it’s going to be a bumpy ride.” Given the way MBAs love to play with math, tampering with the parameters of the unsmoothing function offers lovely possibilities for buffing up risk-adjusted return, while making the results look “all legal.”
Show us da maff!
When you look at how people report performance, there’s often a lot of gaming involved in how they report the IRR
That means if you are not gaming, you lose. Gaming has become the standard.
CalPERS should send Yves a check.
mark to model…
What’s not to like?
These are complicated posts for the uninitiated like me but I may be catching on. I do think it’s a great public service to explain it in a way that beneficiaries can make informed comments to the PTB and wish more savvy readers would be willing to expand upon the arcana
Limited partner investors defend PE general partner valuations – “There is no third party valuation.” (!)
You quote Bob Maynard, CIO Idaho PERS, saying: “So we’re looking for it even if it just gave public market returns, we’d be in favor of it because it has some smoothing effects on both reported and actual risks. …because of its effect having some smoothing of the risk as seen by the accountants and actuaries….”
ah, a fudge factor.
It sounds like the fudge factor itself is the valuable part of PE for pension boards. The classic “Step 3: Then a miracle occurred” in a math equation that won’t work without fudging. That would be understandably attractive to pension boards that can’t make the short term numbers work after the 2008 economic melt down. Longer term, however…. And board members need to be thinking long term, not short term, even if politicians are thinking short term. Sounds like a dilemma for board members. Good intentions aren’t a replacement for sound decisions. If any board members have confused the 2 these reports should help jog their senses. For board members who know the difference and do nothing, well, do they want to be remembered as the Enron guys of Grandma Millie’s pension money?
Thanks again for these great PE pension fund reports.
Sounds like the fudge factor of smoothing is the important part to CalPERS board. If the dilemma is full investment accounting vs. state politicians wanting everything to look great as is then the board is in a tight spot. And that’s assuming the board is willing to honestly assess the predicament, which looks iffy right now. People vested in the pension may have to make a political issue of these investment practices before real change happens. Of course, making a political issue of anything carries a big risk.
Thanks for your great reporting on CalPERS and PE.
The political problem of the need to “smooth” contributions is a significant driver of all this pretending. This might be seen as altruistic, but for the unconscionable rents that are being paid in order to post these fraudulent numbers, and but for the unproductive ways that this taxpayer-generated money is being used as the capital to gut private employment and pension saving.
Maynard: “actual contribution rates we are going to be able to put in place”
I don’t get this. Please tell me Maynard isn’t using happy numbers to induce workers to contribute to his fund.
“designed to validate the status quo” <- That seems uncharacteristically mild, given that the status quo seems to be built on phony accounting.