Risk goes on, Risk goes off

Deutsche Bank’s Stuart Parkinson and Rineesh Bansal kick off their tale of RoRo on a controversial subject: where did the phrase risk-on/ risk-off originate. We’ve had some outlandish claims sent our way in the past.

A colleague who shall remain nameless once claimed he had invented RoRo (and the question mark) but we feel there is a ring of truth to the suggestion that a company called Riskmetrics, originally a JP Morgan project born to gauge the level of risk being run by the firm, actually has the dubious honour.

Following that piece of conjecture (which we are open to correction on), we get a run through of financial crises since the end of the 1980s from Deutsche.

It’s Kindleberger-light and goes through the bond crash of 1994, the Latin American Tequila crisis, continues with those failed geniuses at LTCM, the Asian financial crisis in 1997 and Russia’s subsequent default in 1998.

And 1998, say Parkinson and Bansal, was the first real ‘risk-off’ crisis even it people didn’t really know what that was (our emphasis):

In the weeks following Russia’s unexpected debt default, global debt markets were shook to their core in ways that almost no-one – at the time – could initially understand. Why, for example, was Denmark’s MBS market suddenly in dire-straits? And what on earth were 29.75 year US Treasury bonds doing at such an extraordinary discount to 30 year Treasury bonds? The answer is that it was “Risk-Off” at LTCM (it was lights off shortly thereafter). Suddenly, what seemed previously like a well diversified portfolio of risk positions was correlated to a frightening degree. But for all the bond market carnage, there wasn’t any lasting damage in equity-land. Yes, stocks had plunged by 20% in October, but the S&P500 still ended up on the year.

But not everyone got this “unanticipated correlation” problem:

Despite the market turbulence of the preceding years, for example, regulators (in the 1995 “Amendment to the [Basel] Capital Accord to incorporate market risks”) were giving banks more leeway to rationalise their capital needs based on their own models, having been persuaded by Sir Dennis Weatherstone at JP Morgan and others that internal risk management processes were better equipped to do the job than arbitrary external rules (making Riskmetrics available to the whole marketplace in 1992 was now starting to make some more sense).

So on to 2000/ 2001, Enron and the bursting of the tech bubble. The 47-ish per cent fall in the S&P between 2000 and 2002 was admittedly impressive, but that crisis say Deutsche was an equity crisis as opposed to a debt trauma, and the banking system managed relatively well due in part to rising house prices.

Back to Parkinson’s more personal re-telling:

I’d lived through the 1994 Bond Market Crash, the 1997 Asia Crash, and the 2000 Tech Crash. Markets seemed to be moving progressively more closely together as time passed. In the 1994 Crash, for example, US Equities fell only 10%, whereas in the 1998 Crash, they fell by 20% (remember though, they recovered quickly on both occasions) – in the equity crash of 2000, of course, prices fell by 50% and took years to recover. But then, in 2002, Riskmetrics re-enters the story, completing the theoretical loop between Debt and Equity. As the original 2002 CreditGrades technical document was quick to point out: “…the purpose of the CreditGrades model is to establish a robust but simple framework linking the credit and equity markets”. Little did I know it at the time, but the birth of Creditgrades was also the birth of quantitative Capital Structure Arbitrage (Of course things didn’t always go to plan, as we saw in 2005 when the downgrade of General Motors debt coincided with a bid for the company’s equity by Kirk Kerkorian)

Most of us still didn’t quite know it, but the advent of CreditGrades methodology and its implementation by Capital Structure Arbitrage desks meant that “Risk-On/Risk-Off” now applied to Debt and Equity at the same time. Up until now we’d gotten used to correlation within an asset class (e.g. the contagion risk within Emerging markets in 1997, and between EM and Developed markets in 1998), but not across asset classes. Indeed, the new found correlation between the traditional asset classes was one of the reasons that new, alternative (and historically uncorrelated) asset classes came into their own. And just in case you were wondering how to gain exposure in your investment portfolio to some, livestock, cotton, sugar, natural gas, wine, fine art or just plain-old volatility; along came new forms of investment products to help you out, OEICS, ETFs and 130-30 funds, for example. Most of these new asset classes didn’t yield much – or anything, in cases such as Commodities – but that didn’t matter much at a time when Equities and Debt didn’t yield much either.

For a brief moment prior to the Global Financial Crisis, all asset classes were rising – the old and the new. Commodities, Credit, Emerging Markets, Equity, Fine Art, Government Bonds, High-Yield, Property – all rising. Someone asked me at the time when this had last happened. I looked, but I couldn’t find any precedent. With the benefit of hindsight, this was the moment just before the tidal wave hits when the tide suddenly goes out.

On 25 January 2008, 51 days before Bear Stearns was rescued by JP Morgan, the same Riskmetrics that 20 years ago had given Sir Dennis Weatherstone his first 4:15pm risk report, was IPO’d on the New York Stock Exchange. “Risk- On” had gone public. Pretty soon, we all knew about “Risk-on/Risk-off”.

The big point being made here is that the process of convergence across all markets was a long time coming. This correlated world didn’t sneak up on us and it won’t disappear quickly. It remains to be seen if the “Great Rotation” has the chops to supplant it just yet or ever.

A last piece rather rosy piece from Deutsche:

In 1994, even as Bond yields rose, Equities came out unscathed because “Risk-on/ Risk-off” was still in its infancy. And in the subsequent five years, as the US economy kept growing and Europe did take the next steps, the S&P500 almost doubled, while the Euro-Stoxx 50 index tripled. Perhaps this is helping comfort equity investors today, that as we prepare for Bond yields to rise once again, ‘Risk-on/Risk-off’ seems to be receding from its heady days and is being replaced by ‘The Great Rotation’. After all, the last ‘Great Rotation’ of 1993-4 was good for stocks – so maybe the 2013 Rotation will be equally positive.

It’s a long and recommended read and it resides in the usual place.

Related link:
FT Alphaville’s RoRo file.