Jamie Mai – Seeking Asymmetry

Jamie Mai

This article is part of our 'Guru' series - profiles of successful traders, with takeaways for the UK private investor.
You can find the rest of the series here.

Today’s post is a profile of Guru investor Jamie Mai, who appears in Jack Schwager’s book Hedge Fund Market Wizards. His chapter is called Seeking Asymmetry.

Jamie Mai

Jamie Mai

Jamie Mai features in Michael Lewis’ book The Big Short, which is where Jack Schwager came across him.

Mai’s firm is called Cornwall Capital and Lewis described it as a hedge fund started by dropouts in a shed.

  • In reality Cornwall began as a family office for Mai’s father, who ran AEA Investors, a private equity firm, and before that worked at Lehmans.

After 2011, Cornwall opened up to (a few, sophisticated) outside investors, since Mai regularly came across opportunities that could use more capital than the family office had.

Cornwall uses diverse strategies, but their unifying characteristic is that “they are structured and implemented as highly asymmetric, positive skew trades” – this means that “the upside potential far exceeds the downside risk”.

Cornwall’s short bet on subprime mortgages (which returned 80 times the initial premium they paid for default protection) was what got them featured in Lewis’ book.

Mai also appears in Jack Schwager’s book Hedge Fund Market Wizards.

  • His chapter is called Seeking Asymmetry.

Performance

In the nine years to the time of the interview, Cornwall had returned 40% pa net (52% pa gross).
– Annualised standard deviation is 32% (37% gross).

  • The Sharpe ratio of 1.12 (1.23 gross) is good, but Schwager says this “greatly understates the true return/risk performance of the fund”.

Sharpe uses volatility as a proxy for risk, but because of the design of its trades, most of Cornwall’s volatility is to the upside.

  • They have a volatile pattern of very big gains.

In the real world, investors don’t see upside volatility as a problem.

Schwager uses the “Gain to Pain” ratio instead, which uses a loss-based risk measure.

  • Cornwall’s Gain to Pain ratio of 4.22 (5.2 gross) is exceptionally high.
Early years

Mai did history at college, but was nudged towards accounting by his father, as a backdoor way into private equity.

At the time, NYU had a special master’s of accounting program that was sponsored by the “Big Six” accounting firms who wanted to diversify their staff pool to include undergraduates with liberal arts degrees.

There was a summer crash course to cover four years of accounting classes, and then you were thrown into the financial audit group as a first year staff accountant. I was hired by Ernst & Young.

After finishing the degree, Mai joined Golub Capital, then a small private equity firm (but now a large one).

My experiences in private equity provided me with a solid foundation in fundamental value analysis.

Finding answers is much easier when you know in advance what the questions are. Understanding what the right questions are can be deceptively difficult, but once you do, the rest is straightforward.

Special situations

Mai moved into special situations, which he describes as:

A price dislocation has occurred, either at a single company or across an industry, because the market has identified a particular idiosyncratic risk and assigned an uncertainty discount to it.

Regulation and legal disputes are often involved.

Although markets are good at estimating the magnitude of a contingent liability, they are often poor at evaluating outcomes probabilistically.

Markets tend to over discount the uncertainty related to identified risks and under discount risks that have not yet been expressly identified.

Jamie uses Altria (a tobacco company undergoing litigation) as an example:

The first thing we checked was whether the Altria options still assumed a normal probability distribution, despite the presence of a bimodal event.

Sure enough, they did, which meant the out-of-the-money options were way too cheap.

We bought the out-of-the-money calls and made about 2.5 times our money. But we sold far too soon.

Options

We look for situations where the normal probability distribution assumption implicit in option pricing models is inconsistent with market realities.

Options are priced lowest when recent volatility has been very low. However, the single best predictor of future increases of volatility is low historical volatility.

When volatility gets very low in a market, we start looking for ways to get long volatility, both because volatility is very cheap and because low volatility implies an above-average probability of increased future volatility.

Jamie likes long-dated options:

Often, the longer the duration of the option, the lower the implied volatility, which makes absolutely no sense.

He also likes to exploit discrepancies between forward and spot prices, particularly wide differentials:

Option models generally assume that forward prices are predictive of the future movements in the spot price, but this relationship is often invalid.

Forward option-pricing models can break down, particularly in interest rate markets with steep term structures and low volatility levels.

Risk on, risk off

Another area that Jamie has exploited is:

Forward-looking assumptions based on backward-looking statistical correlations.

After the 2008 crisis (and to an extent, still today), markets have followed a risk on / risk off duality, clearly seen in FX.

In a risk-on environment, everybody piled into currencies with exposure to emerging markets and commodities, such as the Australian dollar.

When it was a risk-off environment, everyone piled into safe-haven currencies, such as the Swiss franc.

Before the crisis, there was no strong correlation between the currencies, but after it, they had an extreme inverse correlation.

At the time, we were looking for an efficient way of getting short the euro. Implied volatility on the euro puts was expensive.

We cheapened the premium substantially by taking our exposure through a worst of option, which is an exotic option that is priced based on a correlation input in addition to the standard inputs for a vanilla option.

Jamie used an instrument with puts on two crosses: euro versus Australian dollar (EUR/AUD) and euro versus Swiss franc (EUR/CHF).

  • “Worst of” are cheaper because the buyer receives payment based on the outcome that is the worst of the two.
  • Because the AUD and CHF were now inversely correlated, this option was very cheap – one or other of the currencies should protect the seller.
See also:  Linda Raschke - The Music of the Markets

But since Jamie was predicting a euro crash, he expected this slide to occur against both currencies.

Using plain vanilla puts, we would have had to pay a premium of about 4½%. But we could do the “worst of” trade for less than one-tenth that amount.

The euro did break down against both the Australian dollar and Swiss franc , and we ended up netting over six times our invested capital.

Schwager comments:

The common denominator in all Jamie’s trades is that markets price securities on the implicit assumption that changes from prevailing levels are equally likely in either direction, whereas in reality, idiosyncratic fundamental factors can make a move in one direction much more likely.

The market assumes symmetry, and he looks for asymmetry.

Volatility

Volatility is a terrible proxy for measuring potential price change over longer intervals of time. For example, if an asset price changes by a constant percentage each day, its volatility will be zero.

The basic concept here is that option prices will tend to be priced too low in smoothly trending markets.

Option-pricing models assume that volatility increases with the square root of time.

Option math works a lot better over short intervals. Once you extend the time horizon, all sorts of exogenous variables are introduced that can throw a wrench into the option-pricing model.

The implication is that long-term options in general will be priced too low.

Capital preservation

The classic value investor achieves capital preservation by taking risks only when he is confident that he won’t lose a meaningful amount of money. We achieve our margin of safety by having a high expected value.

We are comfortable losing 100 percent of our premium four times in a row, as long as we believe that a 25-times payout is likely to occur if we make the same bet 10 times.

Shorting

It is hard to see any circumstances in which we would go outright short a stock.

Your gain is limited, your loss is unlimited, and your exposure grows as you are wrong. Instead, we buy out-of-the-money puts.

When rising implied volatilities make those premiums prohibitively expensive, we shift towards in-the-money puts to reduce the time value decay.

The Big Short

High conviction on an event path priced like a low-probability event is our Holy Grail.”

We got to the trade late. We went from having a probabilistic view that the markets were too confident that home price appreciation would continue indefinitely to having a very high level of conviction that a liquidity bubble existed.

By now, we all know the story of CDOs.

  • subprime mortgages were packaged into asset-backed securities (bonds, known as ABSs or MBSs – mortgage-backed securities).
  • they were stratified by credit risk (AAA, AA), with lower risk bonds being paid off (at a lower interest rate) before the higher risk bonds (at a higher interest rate).
  • below these were “equity” and “mezzanine” tranches that would absorb the first 7% or so of losses and paid even higher interest.

The underlying problem is that by selling on their mortgages to these bonds, the originators passed on the credit risk.

  • So they were incentivised to sell more and more loans, to riskier clients.

As Schwager puts it:

Ultimately, subprime mortgages were being issued with the following characteristics: No down payment. No income, job, or asset verification.

As the risk on the BBB tranches of the bonds rose, they became harder to sell.

So Wall Street revived an idea from the 1990s – the collateralized debt obligation (CDO).

Initially these used diverse forms of credit, but by 2006 they were made of only subprime mortgages.

  • The banks convinced themselves that the top 75% of a security made entirely from BBB loans (ABSs) could be rated AAA.

But the use of entirely BBB loans removed the diversification effect present in the original ABS.

  • These were all loans to risky people, liable to come under repayment pressures under the same economic conditions.

So the 25% buffer of lower rated tranches in the CDO offered only illusory protection.

  • If some of the BBB borrowers defaulted, the chances of more than half of them doing so were high.

There were two other problems:

  1. default rates were based on historical (“real”) mortgages, where the originator cared whether the loan was repaid.
  2. the credit rating agencies were paid by the people selling the CDOs – they shouldn’t have been rating any part of a CDO as AAA.

In 2006, Jamie saw a write-up of a presentation made by Paul Singer of Elliot Associates.

  • Singer showed that the AA tranches of CDOs would fail as soon as house prices stopped rising.

So Cornwall bought credit default swaps (CDSs) on the AA tranches of CDOs.

  • CDSs are effectively insurance on the CDOs defaulting.

CDOs provided us with the opportunity to buy protection on the worst quality bonds at premium levels that were in line with high-grade corporate bonds.

Cornwall hired an analyst to identify the CDOs exposed to the worst ABSs.

The reason we were able to find CDOs whose collateral happened to be the absolute worst-performing subprime MBSs was because [Michael] Burry had gone to dealers months earlier and convinced them to create synthetic securitizations for the specific names he wanted to buy protection on [i.e., to short].

They repackaged those same securities and sold them to us in the form of a synthetic CDO. As far as I know, we were among the first investors to go short CDOs.

We started in October 2006, and the last of our positions was put on in May 2007. The market collapse began on February 1, 2007, when the ABX started tanking.

The ABX is a set of indices of subprime mortgage bonds, classed by credit risk.

Our conviction level spiked up when the ABX tanked and CDO prices didn’t move.

The dealers had bought massive amounts of MBS to hold in inventory in anticipation of turning them into CDOs. They went into overdrive.

Proper mark-to-market prices on the existing CDOs would have killed the CDO market.

The dealers were deliberately mispricing the CDOs.

Our counterparty, Bear Stearns, was marking our CDO positions at cost, even after the ABX fell 30%.

We bought a ton of puts and CDS on Bear Stearns because we thought they would go bankrupt. We went to the SEC.

Jamie is not a fan of the rating agencies, either:

They rated thousands of securities with the same grade as U.S. Treasuries when they weren’t worth more than the paper they were printed on the day they rolled off the press.

I can’t believe they have any credibility left at all.

The CDOs were finally marked down when an index was created of synthetic CDO tranches (the TABX) in early 2007.

At the beginning of August, there was a wave of risk reduction, and everyone wanted to own the CDS protection that we held.

When we liquidated that month, we had positions that were marked at $50,000 on Friday’s close and that we sold for $4.5 million on Monday.

The precipitating event in August was a loss of liquidity in the money markets.

See also:  Steve Watson - Dialling for Dollars
Conclusions

Schwager outlines five principles in Jamie’s strategy:

  1. Find mispricings due to standard pricing assumptions (in derivatives in particular)
  2. Select trades where the probabilities skew to a positive outcome.
    • the expected gain (size x probability) must be twice the expected loss
  3. Asymmetric implementation
    • limited downside, unlimited upside, usually through options
  4. Wait for high-conviction trades.
    • coupled with the 2:1 pay-off ration, this rule generally leads to a concentrated portfolio of 15 to 20 ideas (which might require multiple trades to implement).
  5. Use cash to regulate portfolio risk.
    • option trades don’t tie up much capital, so Jamie normally holds 50% to 80% cash.
    • where he sits in that range determines his portfolio risk level.

Drilling down into options-pricing, Schwager outlines five assumptions that are sometimes wrong:

  1. Prices are normally distributed – future prices near the current level are most probable
    • sometimes large moves are much more likely, and out-of-the-money options will be too cheap.
    • normal distribution also imply that up moves and down moves are equally likely, but this is not alway the case (eg. near lower bounds)
  2. The forward price predicts what will happen to the spot price.
    • this is less likely where the gap between them is large.
    • out of the money options on a price between the forward and the spot will be too cheap.
  3. Volatility scales with the square root of time.
    • this works for short periods, but not longer ones, particularly if current volatility is low.
    • there is more opportunity for mean reversion, and for the development of trends that drive large price moves.
  4. Option pricing models ignore trend.
    • this again implies that out-of-the-money options would be underpriced.
  5. Constant correlations.
    • some market pairs have variable correlations, and if you think a correlation is about to change, the relevant options will be mis-priced.

Schwager also notes Jamie’s flexibility, and his relatively low-risk approach, despite the high volatility caused by the lumpiness of his big wins.

This is a long, dense but ultimately rewarding chapter that focuses on two things:

  1. options mis-pricing
  2. the Big Short during the initial part of the 2008 financial crisis

The first topic is niche and the second is well-known by now (though the explanation in this chapter is particularly clear).

  • So it’s more than possible there will be a limited audience for this interview.

I dallied with options back in the 1980s, but I’m very rusty on the theory these days.

  • They are on my to do list for next year (or the year after).

I like the way that you can use them to refine your risk profile, and in particular to limit downside risk whilst retaining upside potential.

  • Jamie’s comments on (not) shorting stocks are relevant here.

For the average UK private investor, there isn’t too much that can be directly applied to our own portfolios.

  • But it’s a good insight into a central character in the 2008 crisis.

Until next time.

Mike is the owner of 7 Circles, and a private investor living in London. He has been managing his own money for 39 years, with some success.

You may also like...

Leave a Reply

Your email address will not be published.

Jamie Mai – Seeking Asymmetry

by Mike Rawson time to read: 9 min