Financial Econometrics and Empirical Market Microstructure
On Some Approaches to Managing Market Risk Using VaR Limits: A Note
Alexey Lobanov
Abstract Market risk has been traditionally considered in a single-period setting, with fixed positions in a static portfolio and losses caused by price volatility over a specified time horizon. In the real world, however, trading losses are generally a product of both position changes and adverse market movements. Market risk limits have been widely used in the industry for controlling both ex-ante and ex-post losses from traders’ actions, but the interplay of risk limits with risk measurement has been scarcely studied in the literature. This note aims to provide insights into the broad concepts of using limits in market risk management, as well as some approaches to setting and managing market risk limits in a dynamic setting.
Keywords Market risk • Positions limits • Traders’ actions • Trading desk • VaR limits
JEL Classification G21,G32
A common view on market risk presumes that losses are caused by adverse price movements, while positions are fixed over a holding period. For example, in RiskMetrics™ a value-at-risk (VaR) measure is calculated for a static portfolio over a 1-day holding period under the assumption that changes in the portfolio structure and/or composition can be neglected and hence daily P&L is entirely driven by market movements (J. P. Morgan/Reuters 1996).
The views and opinions expressed herein are those of the author and do not necessarily reflect the official position of the Bank of Russia.
A. Lobanov (H)
Banking Regulation Department, Bank of Russia, Moscow, Russia e-mail: alobanov@akado. ru
© Springer International Publishing Switzerland 2015
A. K. Bera et al. (eds.), Financial Econometrics and Empirical Market
Microstructure, DOI 10.1007/978-3-319-09946-0_____ 12
In the real world, however, trading losses are a complex product of both price movements and changes in the portfolio structure.[29] Traders’ actions require revising risk estimated after each material change in the portfolio structure and thus “contaminate” the P&L of a trading desk used in VaR backtesting (Basel Committee on Banking Supervision 1996, 2006). This problem can be mitigated for asset managers and, less effectively, for bank trading desks by checking the results of such “dirty” backtesting, based on theoretical P&L driven only by price movements against “clear” or “cleaned” backtesting[30] (Deutsch 2009). In high - frequency trading, the information value of conventional risk measures, such as VaR, tends to expire, as the holding period may barely exceed a time interval between consequent price movements. For shorter holding periods, traders’ actions come to the forefront as a distinct risk factor, which needs to be understood and managed. Controlling traders’ activity by enforcing position and risk limits should, therefore, be viewed as a primary risk management tool that complements hedging strategies and economic capital.
