FINCAD recently held a webinar, Mastering Negative Rates, to discuss how negative rates came into existence, and offer best practices for modeling them. Guest speakers included Hugh Stewart, Research Director at Chartis Research and Russell Goyder, PhD, Director of Quantitative Research and Development at FINCAD.
Once thought to be impossible, negative rates are now common in many European currencies. Since 2014, the European Central Bank has elected to set the Deposit Facility Rate (DFR) lower than zero in—what some may say— is a desperate attempt to stimulate activity. This phenomenon is presenting challenges for the many firms that are pricing and risk managing derivatives and bonds with models that assume non-negative rate dynamics.
Stewart believes a big problem is that many firms’ models are poorly designed. He backed up his hypothesis with findings from a Chartis Research survey that revealed a need for improved modeling at many firms. Chartis polled 150 senior participants in large organizations on which areas of stress testing they believed need the greatest improvement. 45% pointed to modeling approach as their highest priority.
Stewart likened disorganized models to “messy spaghetti,” when they should be structured and organized. He went on to advise that firms need to find a viable solution for dealing with the negative rates dilemma, as the issue is not expected to disappear anytime soon.
In Goyder’s portion of the presentation, he explained that while code or spreadsheets that reject negative rates are annoying to fix, they are often benign problems. As long as a firm’s underlying calculations still work for negative rates, finding and fixing such bugs should be relatively straightforward. The potentially more dangerous prospect is code or spreadsheets that attempt to correct negative rates for you, by for example, flooring a rate at zero. In this case, it is quite possible that other market participants will exploit your error via a spike in demand for a particular trade before you are even aware of it.
Goyder believes that much of the confusion surrounding negative rates could be resolved by implementing more sophisticated diagnostics such as those accessible in platforms such as FINCAD’s F3. With contextual diagnostics information, firms will be able to understand where specifically to begin to rectify any problems, helping them err on the side of safety. F3 helps firms manage, for example, bonds with negative yields. The platform helps users to view their exposure before and after hedging. It also gives the optimal hedge in order to minimize exposure. Goyder also described that adjusting models to fit negative rates has been a headache for a lot of quants, but it needn’t be this way. He explained how using F3 saves time, as it requires no coding effort when changing models.
Even if firms do have systems with support for rates below zero, there are certain models that simply do not work with negative rates. An example is the Stochastic Alpha Beta Rho (SABR) model used for swaptions. However, to get around this, the general market approach has been to apply a 1% rate shift. Goyder described that despite the need to make tweaks to SABR, it continues to be the best bet to accurately price and hedge derivatives.
Goyder ended his presentation stating that negative rates have caused some stormy waters as of late. To properly respond, financial institutions need to adopt commercially-available solutions that offer contextual diagnostics, flexible curve-building, analytics for handling negative rates and an analytics library able to anticipate OIS discounting. Only with a future-proof enterprise valuation and risk platform will firms be able to weather these and other storms, while retaining a competitive edge.
For more information on the topic of negative rates, see our related blog post, Managing Swaptions in the Face of Negative Rates.