Risk Without Reward

Fair Value Accounting

The Wall Street Journal today carried an opinion piece from Arthur Levitt, the prior head of the SEC, giving a robust defense of the need for mark to market accounting.  Among Levitt's comments:

"The events of the past two weeks constitute the next chapter in a fundamental crisis in financial reporting that has plagued our markets for the past decade. There is a direct line from the implosion of Enron to the fall of Lehman Brothers -- and that's an inability for investors to get sound financial information necessary for making sound investment decisions.

To ask for a suspension in fair-value accounting is to ask the market to suspend its judgment. These trade groups claim that the fair-value accounting standard has distorted banks' balance sheets, and has contributed significantly to the market's volatility.

On the contrary, that gets things backward. It is accounting sleights-of-hand that hid the true risk of assets and liabilities these firms were carrying, distorted the markets, and have caused investors to lose the confidence necessary for our markets to function properly.

Ultimately, those who blame fair-value accounting for the current crisis are guilty of the financial equivalent of shooting the messenger. Fair value does not make markets more volatile; it just makes the risk profile more transparent.

We should be pointing fingers at those at Lehman Brothers, AIG, Fannie Mae, Freddie Mac and other institutions who made poor investment and strategic decisions and took on dangerous risks. Blame should not be paced on the process by which the market learned about them.

We do recognize that bringing the full extent of the risks of these assets and liabilities to light may be a shock to the system. To many this shock may seem unthinkable. But the unthinkable has happened.

It may be painful for some companies, and even for the markets as a whole, as we transition to fair-value accounting. But it is the tough medicine we must take in order to vastly improve financial reporting, bring transparency to the markets, and restore investor confidence."

We couldn't agree more.

www.castlehallalternatives.com

Hedge Fund Operational Due Diligence

Is it a failure of risk management or a failure of compensation?

It doesn't seem like an exaggeration to say that we have just experienced the most tumultuous week in financial market history since 1929.  And this is Monday morning, with another week ahead.

At least financial markets are predictable - once in a millennia events seem to happen every ten years, this being a decade since the LTCM debacle and two decades, give a year, since the '87 crash.

In our view, when we have time to reflect on the causes of this crisis, financial historians (if there are such things) will note that the root cause of these events is simple human nature.  People like to make money.  For some people, though - and investment professionals are likely over-represented in this group - the more money they make, the better they think they are (because if they weren't so good, they wouldn't have made so much money in the first place).  As such, their belief that they are so good makes them want to make several times as much money again.

Bloomberg published an article this week where Jim Grant, the editor of Grant's Interest Rate Observer, was quoted as saying:

``This speaks to a unique and unprecedented calamity of incompetence among people who have been making seven and eight figures each year,'' Grant said in an interview with Bloomberg Television today. ``They ought to consider just what a horror show Wall Street and its facilitators have brought us to. This is a scandal of the first water.''

This morning, the FT also comments on criticism from Gordon Brown, the UK Prime Minister:

“I’ve been saying that when you get your bonuses and your salaries based on short-term deals that have no relationship to long-term performance, then you’ve got to look again at what the system really is doing because if it’s encouraging the short-term fix and the deal rather than making the economy better in the long-run,” he said.

Lord Turner, the FSA’s new chairman, said it was legitimate for regulators to try to prevent investment banks offering staff large pay-outs on profits that were “illusory” and had “simply stocked up problems for the future”. “There are some very important issues to be questioned about the time periods over which bonuses are paid out, the information on which they are measured,” Lord Turner told Sky TV. “These are legitimate issues for regulators to be quizzing banks about.”

The size of bonuses at financial institutions has long been criticised as promoting wealth inequalities in society, but the attacks have acquired a new dimension now that pay structures are seen as contributing to a global financial crisis. Many critics have focused on the “asymmetry” of bonus structures that offer financiers big returns when risky investments succeed but little or no penalty when they fail."

From completely the other end of the political spectrum (and a rather different writing environment) Conrad Black in the Canadian National Post wrote yesterday on "Lessons to learn from Wall Street's week of no return":

"Capitalism does carry the seeds of its own destruction. And odious though government intrusions often are, laissez-faire economics always brings the roof down eventually, as it did in 1929, on what the eminent writer Edmund Wilson rather acidulously called “the pretentious, gigantic, fraud.” That is certainly not being repeated now, but once capitalist zeal can operate in response to financial and imaginative temptation without any even arithmetical oversight, bad things happen, and there is no one but government to provide that oversight. Those who suffer most in these crunches are the victims of collateral damage: shareholders, clerical workers, secondary people who serve the financial machine when it is running at high speed. "

We will all see what this week brings.

www.castlehallalternatives.com

Hedge Fund Operational Due Diligence.

Performance Smoothing

An article in the FT this week commented that "hedge fund investors seek shelter from meltdown."  The discussion touched on the Bear Stearns' funds, raising the broader question of whether investors do - or can - properly adjust hedge fund returns for illiquidity.  In other words, is that nice, smooth "carry trade" return with very low volatility and an exceptionally high Sharpe ratio hiding a risk of significant loss if markets become illiquid? 

One of the other sources quoted by the FT was a study by Riskdata: 

"Olivier le Marois, chief executive of Paris-based Riskdata, a provider of software solutions that helps asset managers and hedge funds control risk, says that given the potential illiquidity of Bear Stearns’ assets, the funds’ steady returns could have raised a red flag. The reason: in a Riskdata study of 3,216 hedge funds and funds of hedge funds, he found that nearly one-third of funds trading illiquid securities may be smoothing their returns, which could mislead investors about actual underlying turbulence."

Riskdata's website has a number of interesting resources, including a more detailed article (see link) called "return smoothing practices: a potential threat to alternative asset growth."

This is very interesting reading, and concludes that procedures should be put in place to "ban" performance smoothing.  To achieve this, the firm makes four recommendations:

"First, thorough due diligence should be made on the valuation process, whether it is performed with in house calculations, vendors software or third parties quotes. Not only should valuations be fair and based on appropriate input data, but all available quotes should be considered without discretionary selection. This point is especially important, due to possible conflicts of interest between managers and brokers about fair valuation when dealing with illiquid securities. 

 

Second, a notable progress in the direction of investors mastering their investments is to give them a quantitative means to detect return smoothing practices. The Bias Ratio – along with Riskdata risk profiling techniques – offers one effective way to alert on possible return smoothing. By itself, quantitative analysis is not sufficient and ought to be supplemented by effective control of managers’ practice. But conversely, quantitative techniques are a necessary add to due diligence and monitoring of investments, as only them can provide an unbiased alert system of managers unfair behaviour. 

 

Third, a complete alarm system should be put in place, which includes not only regular contact with managers, but also a systematic and regular quantitative analysis of their returns, in order to detect biases – like the Bias Ratio – and style drift, as well as to identify sources of risk and potentially hurting scenarios. 

 

Finally, each fund should not only be analysed by itself, but also in the context of a portfolio. For this purpose, the individual risk of one single fund should not hide the risk of diversification – or rather lack of – at the global portfolio level. In other words, this means identifying market scenarios under which investments become correlated and diversification vanishes." 

These remain very sensible and necessary recommendations, and ever more timely as the industry moves towards a new cycle of launching distressed funds trading harder to value securities.

What are some of the lessons to be learned?

1) Investors should always remember the paramount principle of valuation: absent an actual sale, prices used in a NAV calculation are only estimates.  Estimates are, of course, open to manipulation.

2) Best practice universally calls for the valuation process to be segregated from front office investment professionals.  Allowing the front office to price securities creates a clear conflict of interest and an obvious potential for performance smoothing.

3) Investors should take heed of Riskdata's recommendation that all available quotes are included in the pricing process.

4) Very careful due diligence is always required to evaluate valuation procedures adopted by the manager as well as the administrator.  In Castle Hall's experience across the industry, we continue to see enormous variation in valuation procedures.  Investors need to understand, in detail, the valuation process adopted by each fund: investors should not assume that large funds always have good valuation procedures and, as we have commented frequently, should definitely not assume that the administrator is actually pricing the portfolio independently.  Investors must also be conscious that funds which adopt different valuation procedures may not produce directly comparable NAVs.

As always, caveat emptor.

www.castlehallalternatives.com

Hedge fund operational due diligence

Is 2 and 20 fundamentally flawed?

Two articles this morning highlight the challenges of running a hedge fund in difficult markets.

Reuters comments that "exiting hedge fund managers cast pall over hedge fund industry".  Discussing the recent closures of Dan Benton's Andor Capital, Ron Insana (which we discussed here) and Turnberry Capital, Reuters notes that running a hedge fund is "no longer a sure thing":

"Hedge fund managers often promised to make money in all markets but several said that shorting stocks, the way to make money in down markets, is becoming more difficult and expensive as ever more investors are trying that strategy, making it tougher and costlier to locate the stocks to short. In turn, funds' potential for profits are reduced as their bets are no longer a sure thing.

This means the prospect of earning a 20 percent performance fee on top of a 2 percent management fee, numbers that lured thousands of traders and portfolio managers into the industry, is in jeopardy.

Performance fees are paid for gains, not losses, and this year some individual hedge funds have lost as much as 20 percent, some investors who saw the data said.

"By my math, some people are taking a pretty big pay cut to be running a hedge fund and who needs that?" said one manager who asked to remain anonymous in order to speak candidly."

Separately, the FT comments on the poor performance of the Atticus funds (Atticus Europe -33%, Atticus Global -25%).  What caught our eye in this article was the final paragraph:

"Rob Coburn, head of investor relations at Atticus, said the fund was a long-term investor. "While year-to-date performance has been disappointing it is too short a period of time to measure our investment performance," he said. "We believe a better measure is our longer-term record of two, three or five years.""

How do these comments link together?  They highlight - once more - the potential pitfalls of 2 and 20, whereby hedge fund managers are paid on the increase in the paper value of their portfolio once per year (or, for some funds with even more aggressive terms, semi annually, quarterly or even monthly.)

We have discussed the pros and cons of the hedge fund compensation scheme before and, as we watch the markets evolve, our thinking becomes ever firmer that 2 and 20 (or 1 and 20 through to 4 and 44, pick a number) may be fundamentally flawed.

The problems are simple:

1) Hedge fund managers receive an incentive fee on any positive performance, not just performance above the market (i.e. for all the talk of alpha, managers get paid on beta as well.)

2) There is a frequent mis-alignment between the annual cycle for incentive compensation and the investment time horizon, which is often considerably longer than 12 months.

3) Unlike private equity managers, hedge funds pay themselves incentive fee compensation on unrealized appreciation - paper profits - as well as realized gains.  

What are the lessons?

Taking the first point, it would seem to us that a hurdle rate should be applied to any performance fee.  If this industry truly is about alpha, incentive fees should be paid on out-performance, not just performance.  Yet, rare is the fund that adds a LIBOR hurdle, for example, to the performance fee calculation.

On the second point, if a hedge fund has an investment time horizon longer than one year, the incentive fee should be calculated - and paid - over a matching, multi-year period.  At a very minimum, in our view, incentive fees should not be calculated over a period shorter than the investors' lock up: if you're locked up for three years, for example, the manager should not receive any incentive fee until the three year mark, which is the first opportunity for investors to decide whether they want to retain capital with the manager.

More generally, the incentive fee calculation period should be in sync with the advertised investment time horizon and portfolio "cycle".  Taking the example of Atticus above, why is it that a firm which believes a "better measure [of performance] is two, three or five years" still pays itself every year, beyond the reason that this has become industry convention?

The final issue of realized versus unrealized gains is equally significant.  This is not a direct problem for long short equity and CTA funds which trade exchange traded (level 1) instruments, but can become a huge issue for funds trading highly complex, hard to value paper.  

Firstly, the managers may well be pricing these positions themselves, leading to a blatant and entirely inappropriate conflict of interest.  As we have discussed frequently, investors should not assume that third party administration companies, when appointed, are independently pricing the tricky securities which do have pricing risk - the admins are too happy to price the easy stuff but take manager marks for the positions which actually are hard to value.

Secondly, some hedge funds hold instruments which unavoidably are hard to price: especially in dislocated markets, there may be insufficient price transparency and market depth for mark to market levels to be reliably used as the basis of compensation.  If a fund is trading highly illiquid assets, the only true test of value is an actual sale, which leads us to conclude that sale prices, not month to month price estimates, should be used as the basis for fee calculations.  This, of course, opens a broader question of whether some hedge funds hold assets which are so illiquid and hard to price that they are profoundly unsuitable for an open ended fund structure in the first place, but that is a topic for another post.

Going forward, the incentive fee structure will need to evolve.  There is no question that investors will pay for genuine outperformance, but the current, annual incentive fee creates too many instances where manager and investor interests can diverge.  If the hedge fund industry is to flourish over the long term, through bad markets as well as good, that alignment of interest needs to be strengthened - hedge funds cannot assume that investors will always continue to add new capital, no questions asked.  

We also believe that moving incentive fee compensation to a multi-year time horizon would positively impact the psychology of the industry.  At present, the annual fee cycle is too close to the model of a prop desk, whereby traders don't have any sense of time beyond their next bonus check (traders must be the only people with a time horizon shorter than a politician!)  Building a successful investment management business, however, requires phenomenal commitment and an ability to endure adverse markets just as much as it involves enjoying the benefits when markets are strong.  It seems, though, that plenty of new entrants to the hedge fund industry have a baseline assumption that "average" or "normal" compensation includes the incentive fees generated from a very good year.  In reality, not every year will be good: as a result, not every trader will make millions every year.  Again, the industry stands a better chance of long term success if there is less focus on short term compensation.

www.castlehallalternatives.com

Hedge fund operational due diligence