Recently, I have focused on risk production and control.
Reading through the various regulations and proposals, errors, in the market and credit risk production itself, fall into, and get a little lost within, the op risk side of life. My main issue with this is that far from being an economic capital driver, good risk production control is the central plank in establishing an adoptable risk culture in any firm.
My main fear is that the general, and understandable, focus on meeting the regulatory changes has been at the cost of moving forward with better risk controls.
Effectively, if the lack of risk as a cultural imperative within financial firms is a major contributor to the recent crisis, then its correction should be a major driver in the aftermath. It is difficult to see how there can be any serious adoption, or buy in, from the stakeholders of risk, if the numbers themselves cannot be trusted, or at least correct quickly, once errors have been detected.
From what I have read, most of the control functions are around the trading itself, with risk incentives being in the form of punative measures such as CVA or IRC (sticks rather than carrots). On the risk side itself, it has all been about the capital numbers (again, sticks over carrots), with CVA, IRC, Stressed VaR, DSR etc.
I have always felt that making the risk numbers useful from a strategic point of view, which means that they are genuinely useful in terms of planning, would put risk at the heart of a firm in a far more 'sticky way' than simply making it a capital issue (which leads, in part, to increasing product sophistication and model/complexity arbitrage). It does require, however, that the numbers are either correct, or quickly correctable, and therefore trusted.
As I said, this has been a main focus for me and my firm for a while. i would love to hear other folks opinions.
Replies
CVA, IRC, VaR, DSR and a description of the roles they play in an organizations finances and management.
- culture adoption, of risk management, of production control as "Core", and
- capital issue as "Periphery".
Both are Insight-Building Models based on past, present and extrapolation of the future, for which Core relates to macroeconomic and microeconomic (but I'm limiting my discussion to macro only) whilst Periphery relates to sub-problems of the macroeconomic.
a) The Problem with Core :-
We are too far from absolute truth to be so specialised and to make the kind of confident quantitative claims that often emerge from Core since it is built on supposedly "micro-founded" calibrations of key parameters that are definitely on the surreal side ie by not taking Periphery considerations, the model narrows the parameter to certain values. This is notwithstanding the quantitative mathematical formalisations of Core where what's mathematically equivalent may not be psychologically so.
b) Policy Implications of Confusing Environment :-
How do we go about doing policy analysis in models with some loosely specified blocks not pinned down to specific areas of our interest eg the centrality of surprises in financial and economic crises seems discouraging since it would seem that it is difficult to fight something that is essentially impossible to predict, that keeps changing, and that is not understood until after it happens thereby giving rise to improvise policy.
c) The Problem with Periphery :-
Is not the model itself (VAR, CVAR, Stressed VAR, GARCH etc etc), it is Data Reliability (DR). Without DR, it is literally impossible to work !!! Risk data are sitting in a number of silos across the globe without any viable way of aggregating them in a useful way. This data includes "monopoly" info from key central banks that needs to be integrated and managed along with all financial institutions including hedge funds.
- unless the safety net is backed by solid crisis planning, cumulative extensions of the safety net are apt to result in less frequent but more devastating crises,
- the more effective a nation's safety net becomes, the less likely that regulatory personnel will have prior hands-on experience in coping with the severity of crisis pressures, and
- redesigning regulatory schemes and relocating bureaucratic responsibilities for different features of the safety net will not by itself do much to slow processes of regulatory arbitrage.
a) globalisation is dead or "prison" economy,
b) markets are deeply embedded in systems of governance, and
c) democratic governance organises within national political communities eg EU.
Welcome to the Real World.