Financial risk management

The 5% Value at Risk of a hypothetical profit-and-loss probability density function

Banks and other wholesale institutions face various financial risks in conducting their business, and how well these risks are managed and understood is a key driver behind profitability, as well as of the quantum of capital they are required to hold.[22] Financial risk management in banking has thus grown markedly in importance since the Financial crisis of 2007–2008. [23] (This has given rise [23] to dedicated degrees and professional certifications.)

The major focus here is on credit and market risk, and especially through regulatory capital, includes operational risk. Credit risk is inherent in the business of banking, but additionally, these institutions are exposed to counterparty credit risk. Both are to some extent offset by margining and collateral; and the management is of the net-position. Large banks are also exposed to Macroeconomic systematic risk - risks related to the aggregate economy the bank is operating in[24] (see Too big to fail).

The discipline [25][26] [7][8] is, as outlined, simultaneously concerned with (i) managing, and as necessary hedging, the various positions held by the institution — both trading positions and long term exposures; and (ii) calculating and monitoring the resultant economic capital, as well as the regulatory capital under Basel III - with the latter as a floor. The calculations here are mathematically sophisticated, and within the domain of quantitative finance.

Broadly, calculations [26][25] are built for (i) on the "Greeks", the sensitivity of the price of a derivative to a change in its underlying parameters, as well as on the various other measures of exposure to market factors, such as DV01 for the sensitivity of a bond or swap to interest rates; and for (ii) on value at risk, or "VaR", an estimate of how much the investment or area in question might lose with a given probability in a set time period, with the bank holding economic “risk capital” correspondingly.

The regulatory capital quantum is calculated via specified formulae: risk weighting the exposures per highly standardized asset-categorizations, applying the aside frameworks, and the resultant capital - at least 12.9% of these Risk-weighted assets - must then be held in specific "tiers" and is measured correspondingly. In certain cases, banks are allowed to use their own estimated risk parameters here; these "internal ratings-based models" typically result in less required capital, but at the same time are subject to strict minimum conditions and disclosure requirements.

The financial crisis exposed holes in the mechanisms used for hedging; see Fundamental Review of the Trading Book § Background, Tail risk § Role of the global financial crisis (2007-2008), Value at risk § Criticism, and Basel III § Criticism. As such, the methodologies employed have had to evolve:

• A core technique continues to be Value at Risk — applying the traditional parametric and "Historical" approaches — but now supplemented [26] with the more sophisticated Conditional value at risk / expected shortfall, Tail value at risk, and Extreme value theory (and PFE and EE for regulatory). For the underlying mathematics, these may utilize mixture models, PCA, volatility clustering, copulas, and other techniques.[27]
• For the daily direct analysis of the positions at the desk level, as a standard, measurement of the Greeks now inheres the volatility surface — through local- or stochastic volatility models — while re interest rates, discounting and analytics are under a "multi-curve framework". Derivative pricing now embeds counterparty and other considerations [28] through the CVA and XVA "valuation adjustments".
• Additional to these,[26] are various forms of stress test[29] and scenario analytics, and related economic capital optimization. These tests are typically linked to the macroeconomics, and provide an indicator of how sensitive the bank is to changes in economic conditions, and of its ability to respond to market events. And here, more generally, providing estimates for scenarios beyond the VaR thresholds, thus “preparing for anything that might happen,” rather than worrying about precise likelihoods.[30]
• Model risk is addressed [31] through regular validation of the models used by the bank's various divisions; for VaR models, backtesting.
• Under what is sometimes called "Basel IV", several regulatory capital standards have been modified, in particular via FRTB and SA-CCR; other modifications are being phased in from 2023.

Re implementation, Investment banks, particularly, employ dedicated "Risk Groups", i.e. Middle office teams monitoring the firm's risk exposure to, and the profitability and structure of, its various businesses, products, asset classes, desks, and / or geographies.[32] By increasing order of aggregation: (i) Financial institutions will typically [25] set limit values for each of the Greeks, or other measures, that their traders must not exceed, and traders will then hedge, offset, or reduce periodically if not daily - see the techniques listed below. (ii) Desks, or areas, will similarly be limited as to their VaR quantum (total or incremental, and under various calculation regimes), corresponding to their allocated economic capital; a loss which exceeds the VaR threshold is termed a "VaR breach". Allocated regulatory capital is similarly monitored. (iii) Each area's (or desk's) concentration risk will be checked [33][32][34] against thresholds set for various types of risk, and / or re a single counterparty, sector or geography. (iv) Leverage will be monitored - at very least re regulatory requirements - as leveraged positions could lose large amounts for a relatively small move in the price of the underlying. (v) Periodically,[35] these all are estimated under a given stress scenario, and risk capital - together with these limits - [36] is correspondingly revisited.

A key practice, [37] overlapping and assimilating these, is to assess the Risk-adjusted return on capital, RAROC, of each area (or product). Here,[38] the (ex-post) economic return is divided by allocated-capital; and this result is then compared [38][22] to the target-return for the area — usually, at least the equity holders' expected returns on the bank stock [38] — and identified under-performance can then be addressed. (See similar below re. DuPont analysis.) The numerator here is calculated as achieved trading-return, less a risk- and term-appropriate funding charge determined under the bank's funds transfer pricing (FTP) framework; direct costs are sometimes also subtracted. The denominator is the area's allocated capital, as above, increasing as a function of position risk. [39] [40] [37]

Other teams, overlapping the above Groups, are then also involved in risk management. In their Front office, Banks employ specialized XVA-desks tasked with centrally monitoring and managing their CVA and XVA exposure, typically with oversight from the appropriate Group.[28] Middle office maintains the following functions also: Product Control is primarily responsible for insuring traders mark their books to fair value (a key protection against rogue traders) and for "explaining" the daily P&L. Credit Risk monitors the bank's debt-clients on an ongoing basis, re both exposure and performance. Corporate Treasury is (co)responsible for maintaining the FTP framework, and for monitoring overall funding, capital structure, and liquidity risk (reviewing i.a. Liquidity at risk, Earnings at risk, and Cash flow at risk).

Performing the above tasks — while simultaneously ensuring that computations are consistent over the various areas, products, teams, and measures — requires that banks maintain a significant investment in sophisticated infrastructure, finance / risk software (the quantitative models, typically built in-house), and dedicated staff. Risk software often deployed is from FIS, Kamakura / SAS, Murex, and Numerix.

Contribution analytics: Profit and Loss for units sold at current fixed costs.

The same, for varying (scenario-based) Revenue levels, at current Fixed and Total costs.

In corporate finance and financial management, [41] [10] financial risk management, as above, is concerned more generally with business risk - risks to the business’ value, within the context of its business strategy and capital structure. [42] The scope here - ie in non-financial firms [12] - is thus broadened [9] [43] [44] (re banking) to overlap enterprise risk management, and financial risk management then addresses risks to the firm's overall strategic objectives, incorporating various (all) financial aspects [45] of the exposures and opportunities arising from business decisions, and their link to the firm’s appetite for risk, as well as their impact on share price. In many organizations, risk executives are therefore involved in strategy formulation: "the choice of which risks to undertake through the allocation of its scarce resources is the key tool available to management." [46]

Re the standard framework,[45] then, the discipline largely focuses on operations, i.e. business risk, as outlined. Here, the management is ongoing [10] — see following description — and is coupled with the use of insurance, [47] managing the net-exposure as above: credit risk is usually addressed via provisioning and credit insurance; likewise, where this treatment is deemed appropriate, specifically identified operational risks are also insured.[44] Market risk, in this context,[12] is concerned mainly with changes in commodity prices, interest rates, and foreign exchange rates, and any adverse impact due to these on cash flow and profitability, and hence share price.

Correspondingly, the practice here covers two perspectives; these are shared with corporate finance more generally:

1. Both risk management and corporate finance share the goal of enhancing, or at least preserving, firm value.[41] Here,[9][45] businesses devote much time and effort to (short term) liquidity-, cash flow- and performance monitoring, and Risk Management then also overlaps cash- and treasury management, especially as impacted by capital and funding as above. More specifically re business-operations, management emphasizes their break even dynamics, contribution margin and operating leverage, and the corresponding monitoring and management of revenue, of costs, and of other budget elements. The DuPont analysis entails a "decomposition" of the firm's return on equity, ROE, allowing management to identify and address specific areas of concern,[48] preempting any underperformance vs shareholders' required return.[49] In larger firms, specialist Risk Analysts complement this work with model-based analytics more broadly;[50][51] in some cases, employing sophisticated stochastic models,[51][52] in, for example, financing activity prediction problems, and for risk analysis ahead of a major investment.
2. Firm exposure to long term market (and business) risk is a direct result of previous capital investment decisions. Where applicable here [12][45][41] — usually in large corporates and under guidance from [53] their investment bankers — risk analysts will manage and hedge [47] their exposures using traded financial instruments to create commodity-,[54][55] interest rate-[56][57] and foreign exchange hedges [58][59] (see further below). Because company specific, "over-the-counter" (OTC) contracts tend to be costly to create and monitor — i.e. using financial engineering and / or structured products — ”standard” derivatives that trade on well-established exchanges are often preferred. [14][45] These comprise options, futures, forwards, and swaps; the "second generation" exotic derivatives usually trade OTC. Complementary to this hedging, periodically, Treasury may also adjust the capital structure, reducing financial leverage - i.e. debt-funding - so as to accommodate increased business risk (they may also suspend dividends [60]).

Multinational corporations are faced with additional challenges, particularly as relates to foreign exchange risk, and the scope of financial risk management modifies significantly in the international realm.[58] Here, dependent on time horizon and risk sub-type — transactions exposure[61] (essentially that discussed above), accounting exposure,[62] and economic exposure [63] — so the corporate will manage its risk differently. Note that the forex risk-management discussed here and above, is additional to the per transaction "forward cover" that importers and exporters purchase from their bank (alongside other trade finance mechanisms).

It is common for large corporations to have dedicated risk management teams — typically within FP&A or corporate treasury — reporting to the CRO; often these overlap the internal audit function (see Three lines of defence). For small firms, it is impractical to have a formal risk management function, but these typically apply the above practices, at least the first set, informally, as part of the financial management function; see discussion under Financial analyst. Correspondingly, the discipline relies on a range of software,[64] from spreadsheets (invariably as a starting point, and frequently in total [65]) through commercial EPM and BI tools, often BusinessObjects (SAP), OBI EE (Oracle), Cognos (IBM), and Power BI (Microsoft).

Hedging-related transactions will attract their own accounting treatment, and corporates (and banks) may then require changes to systems, processes and documentation; [66] [67] see Hedge accounting, Mark-to-market accounting, Hedge relationship (finance), Cash flow hedge, IFRS 7, IFRS 9, FASB 133, IAS 39, FAS 130.

Efficient Frontier. The hyperbola is sometimes referred to as the "Markowitz bullet", and its upward sloped portion is the efficient frontier if no risk-free asset is available. With a risk-free asset, the straight capital allocation line is the efficient frontier.

Here maximizing return and minimizing risk such that the portfolio is Pareto efficient (Pareto-optimal points in red).

Fund managers, classically,[68] define the risk of a portfolio as its variance[11] (or standard deviation), and through diversification the portfolio is optimized so as to achieve the lowest risk for a given targeted return, or equivalently the highest return for a given level of risk; these risk-efficient portfolios form the "Efficient frontier" (see Markowitz model). The logic here is that returns from different assets are highly unlikely to be perfectly correlated, and in fact the correlation may sometimes be negative. In this way, market risk particularly, and other financial risks such as inflation risk (see below) can at least partially be moderated by forms of diversification.

A key issue in diversification, however, is that the (assumed) relationships are (implicitly) forward looking. As observed in the late-2000s recession historic relationships can break down, resulting in losses to market participants believing that diversification would provide sufficient protection (in that market, including funds that had been explicitly set up to avoid being affected in this way[69]). A related issue is that diversification has costs: as correlations are not constant it may be necessary to regularly rebalance the portfolio, incurring transaction costs, negatively impacting investment performance;[70] and as the fund manager diversifies, so this problem compounds (and a large fund may also exert market impact). See Modern portfolio theory § Criticisms.

Addressing these issues, more sophisticated approaches have been developed in recent times, both to defining risk, and to the optimization itself. (Respective examples: (tail) risk parity, focuses on allocation of risk, rather than allocation of capital; the Black–Litterman model modifies the "Markowitz optimization", to incorporate the views of the portfolio manager. [71]) Relatedly, modern financial risk modeling employs a variety of techniques — including value at risk, historical simulation, stress tests, and extreme value theory — to analyze the portfolio and to forecast the likely losses incurred for a variety of risks and scenarios. In parallel, [72] managers - active and passive - also seek to understand any tracking error, i.e. underperformance vs a "benchmark", and here often use attribution analysis preemptively so as to diagnose the source early, and to take corrective action (see also Fixed-income attribution); they similarly use style analysis to address style drift. Fund Managers typically rely on sophisticated software here (as do banks, above); widely used platforms are provided by BlackRock, Eikon, Finastra, Murex, Numerix, MPI and Morningstar.

Additional to these (improved) diversification and optimization measures, and given these analytics, Fund Managers will apply specific risk hedging techniques as appropriate;[68] [11] these may relate to the portfolio as a whole or to individual holdings.

• Fund managers may engage in portfolio insurance, a hedging strategy developed to limit the losses an investor might face from a declining index of stocks without having to sell the stocks themselves. This strategy involves selling Stock market index futures during periods of price declines. The proceeds from the sale of the futures help to offset paper losses of the owned portfolio. Alternatively, and more commonly,[73] they will buy a put on a Stock market index option so as to hedge. In both cases the logic is that the (diversified) portfolio is likely highly correlated with the stock index it is part of; thus if stock prices decline, the larger index will likewise decline, and the derivative holder will profit.[74] (Inflation, which affects all securities, [75] can to some extent be hedged using inflation-linked bonds;[76] diversification here is achieved by including tangible assets and commodities in the portfolio.[77])
• Fund managers, or traders, may also wish to hedge a specific stock's price. Here, they may likewise [74] buy a single-stock put, or sell a single-stock future. Alternative strategies may rely on assumed relationships between related stocks, employing, for example, a "Long/short" strategy.
• Bond portfolios, when e.g. a component of an Asset-allocation fund or other diversified portfolio, are typically managed similar to equity above: the Fund Manager will hedge her bond allocation with bond index futures or options.[78][79][74] In other contexts, the concern may be the net-obligation or net-cashflow. Here the fund manager employs Interest rate immunization or cashflow matching. Immunization is a strategy that ensures that a change in interest rates will not affect the value of a fixed-income portfolio (an increase in rates results in a decreased instrument value). It is often used to ensure that the value of a pension fund's assets (or an asset manager's fund) increase or decrease in an exactly opposite fashion to their liabilities, thus leaving the value of the pension fund's surplus (or firm's equity) unchanged, regardless of changes in the interest rate. Cashflow matching is similarly a process of hedging in which a company or other entity matches its cash outflows - i.e., financial obligations - with its cash inflows over a given time horizon.
• For individual bonds and other fixed income securities, specific credit and interest rate risks can be hedged using interest rate- and credit derivatives. Sensitivities re interest rates are measured using duration and convexity for bonds, and DV01 and key rate durations generally, and an offsetting derivative-position is purchased. For credit risk, analysts use models such as Jarrow–Turnbull and KMV to estimate the probability of default, hedging where appropriate. (At a portfolio level — often for credit-VaR — they may also use a transition matrix of Bond credit ratings [80] to estimate the probability and impact of a "credit migration",[81][82] aggregating the bond-by-bond result.) Interest rate- and credit risk together, may be hedged via a Total return swap. See Fixed income analysis
• For derivative portfolios, and positions, the Greeks are a vital risk management tool: as above, these measure sensitivity to a small change in a given underlying price, rate, or parameter, and the portfolio is then rebalanced accordingly[74] by including additional derivatives with offsetting characteristics, or by purchasing or selling specified units of the underlying security.