Best tech vendor for ESG: Conning

Tech vendor upgrades economic scenario generator for the new normal

hal-pedersen
Hal Pedersen, Conning. Image: Carrie Draghi Photography

Buy-side Awards 2016

As asset managers have slowly come to realise low interest rates and slow growth are not just a passing phase, they have been looking at how they can adjust their market modelling and investment strategies to take the new normal into account.

The first port of call for many has been their economic scenario generator (ESG). Because an ESG generates the financial and macroeconomic variables that underpin risk modelling, it is here that investment managers want to see their views on future market dynamics reflected. Conning has responded by allowing users greater ability to recalibrate its Gems ESG software to explore such views.

"Clients aren't certain where all of their risks may come from, but they know they can vary a lot. They want to impose their own views on various aspects of model performance so they can test their concerns," says Hal Pedersen, a director of risk solutions at Conning, based in Hartford, Connecticut.

The company has added to its real-world calibration tool the ability to simultaneously specify investment targets at time zero and multiple future time horizons, as well as by multiple tenors, and also the ability to simultaneously target a wide range of statistical measures, such as mean, standard deviation, percentiles and skewness. Typical investment targets for calibrating the ESG include levels or volatilities for interest rates, expected returns and return volatility, and dividend yields.

"Rather than clients trying to pick the right long-term target, we enable them to do a robust combination of calibrations: where in one set of scenarios, they may set a higher target with slower mean reversion: while in another set of scenarios, they may go for a lower target with more rapid mean reversion," says Pedersen.

Rather than clients trying to pick the right long-term target, we enable them to do a robust combination of calibrations: where in one set of scenarios, they may set a higher target with slower mean reversion: while in another set of scenarios, they may go for a lower target with more rapid mean reversion
Hal Pedersen, Conning

Conning has also been enhancing its ESG to capture the ultra-low interest rate behaviour that has become familiar over the past decade. "Traditional interest rate models are a challenge in the current market, where the short rate is pinned to zero and the long rate is moving around. So we have been developing some new dynamics whereby you can capture circumstances where you have more volatility in the longer maturity than you do in the shorter maturity," says Pedersen.

At the same time, Conning has responded to clients' concerns about potential credit risk scenarios as the possibility increases of interest rates normalising, as well as the market turbulence that such normalisation would likely entail.

It has added a contagion effect to its sovereign debt model that allows for the modelling of multiple sovereign issuers in any currency zone, and provides for a contagion process that will make credit events more likely to have a delayed domino effect. Although the enhancement was added to capture the possibility of the problems of Greece or other weaker countries spreading across the eurozone, it could also potentially be used to model the impact of the UK's Brexit vote.

"We haven't parameterised the model for a 'hard Brexit' factor yet, but if you elevate the initial conditions of the state variables that drive the model, then you can get the effect," says Pedersen.

Conning has also recently added a feature to its ESG to recognise credit ratings drift. "We allow the fluctuation where on average you maintain a rating with a 90% probability, but the transition behaviour in the model can drift around stochastically. This gives clients who want to test a range of investment programmes the opportunity to create more stress scenarios that capture some of the rating migration patterns we saw post-crisis," says Pedersen.

Continental shift

The largest portion of Conning's ESG clients are general insurers, particularly in the US, but the company also has a growing number of life insurance clients in Europe and Asia.

European clients include Talanx, which uses Gems to support its Solvency II internal model that was recently approved by the German regulator. For the Asian market, Conning has added support for cross-currency swaps and functionality to model the peculiarities of China's financial bonds, as well as a corporate bond model for less liquid markets, such as Malaysia.

A number of the new features that Conning has been introducing to give more accurate and insightful risk statistics are numerically intensive. To support the processing demands, the company is introducing a new architecture that takes advantage of emerging programming languages designed specifically for distributed parallel processing that can speed up calculations by a factor of five or more, as well as offering the option of scalable cloud computing.

Alongside this, it has also improved the speed of recalibration of its ESG and the automation of modelling processes. Conning is planning to introduce a new, more familiar and accessible user interface early in 2017, which will allow a broader range of people to interpret results, and will continue to enhance recalibration capabilities for its Gems ESG software.

"A robust ESG allows an insurer to impose a view while also having a coherent set of risk scenarios that provide insight into its business across a wide range of possible outcomes. It should produce some extreme but plausible scenarios that can reveal unexpected exposures, and provide a flexible calibration range while still imposing an overall benchmark on the risk scenarios," says Pedersen.

  • LinkedIn  
  • Save this article
  • Print this page  

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: