UBS AG
Versione ottimizzata per screen reader per visitatori ipovedenti e non vedenti Home | Accessibilità | Versione zoomata | Sitemap locale | Service Finder | eng deu fra ita | Cerca
   
Relazioni 2007  
Strategy, Perf. & Resp. (ING) Risk, Treas. & Cap. Mgmt (ING) CG & Comp. Report (ING) Fin. Stat. (ING) Retrospettiva
     
Risk management
Treasury and capital management
More about UBS
 

Market risk
Market risk

Market risk is the risk of loss from changes in market variables. There are two broad categories of variables – general market risk factors and idiosyncratic components. General market risk factors are variables which are driven by macroeconomic, geopolitical and other market-wide considerations, independent of any instrument or single name. They include the level, slope or shape of yield curves (interest rates), the levels of equity market indices and exchange rates, prices of energy, metals and commodities, and the general level of credit spreads – the yield paid by borrowers above that on risk-free securities. Associated volatilities and correlations between risks factors – which may be unobservable or only indirectly observable – are also considered to be general market risk factors. Idiosyncratic components are those that cannot be explained by general market moves – broadly, elements of the prices of debt and equity instruments and derivatives (including derivative securities and basket products) linked to them, that result from factors and events specific to individual names.

UBS discloses its market risk in terms of statistical loss using its proprietary Value at Risk ("VaR") model, but internally also applies stress measures and a variety of concentration and other quantitative and qualitative controls.

In 2007, the market for US residential mortgage-related products – a previously large and highly liquid market – suffered extreme moves and severe dislocation. Although, conceptually, the market risk framework includes appropriate types of controls, their detailed implementation did not cover all the dimensions of risk measurement and aggregation which, in the extreme events of 2007, proved to be necessary. The Investment Bank had accumulated positions, predominantly in highly rated instruments, which were not identified as concentrations. Enhancements have been made and will continue to be made to reflect the lessons learnt.

Sources of market risk

UBS takes both general and idiosyncratic market risks in its trading activities, and some non-trading businesses create general market risks.

Trading

Most of UBS's trading activity is in the Investment Bank. It includes market-making, facilitation of client business and proprietary position taking in the cash and derivative markets for equities, fixed income, interest rates, foreign exchange, energy, metals and commodities.

The fixed income trading area of fixed income, currencies and commodities (FICC) carries inventory, including exposures to residential and commercial real estate, corporate and consumer credit, and US municipal and student loan markets. Credit spread exposure from FICC positions is generally the largest contributor to VaR.

Exposure to movements in the level and shape of yield curves arises in all the Investment Bank's trading activities but predominantly in parts of FICC. Exposure to directional interest rate movements varies depending on client flows and traders' views of the markets. It is often these variations that drive changes in the level of Investment Bank VaR, although the impact of any position or change in position depends on the composition of the whole portfolio at the time.

UBS is active in all major equity markets and an increasing number of newer markets. Equity risk is the other major contributor to Investment Bank market risk, partly from index-based transactions but also from individual stocks, giving rise to idiosyncratic as well as general market risk. A significant component of equity VaR is event risk from proprietary positions, which are taken, for example, to capture arbitrage opportunities or price movements resulting from mergers and acquisitions.

UBS trades in large volumes in currencies, and to a lesser extent in energy, metals and commodities, but the contribution from these activities to overall market risk has generally been relatively small.

The asset management and wealth management businesses carry small trading positions, principally to support client activity. The market risk from these positions is not material to UBS as a whole.

Trading businesses are subject to a variety of market risk limits within which traders manage their risks according to their view of the market, employing a variety of hedging and risk mitigation strategies. Senior management and risk controllers may, however, give instructions for risk to be reduced, even when limits are not exceeded, if particular positions or the general levels of exposure are considered inappropriate. Hedging and mitigation strategies are then discussed and agreed with trading management.

Non-trading

In the Investment Bank, significant non-trading interest rate risk and all non-trading foreign exchange risks are captured, controlled and reported under the same risk management and control framework as trading risk.

In the other business groups, exposures to general market risk factors – primarily interest rates and exchange rates – also arise from non-trading activities. Market risks are generally transferred to the Investment Bank or Treasury, who manage the positions as part of their overall portfolios within their allocated limits. The largest items are the interest rate risks in Global Wealth Management & Business Banking. All risk transfers take place according to approved transfer pricing mechanisms. Market risks that are retained by the other business groups are not significant relative to UBS's overall risk, and all exposures are subject to market risk measures and controls. With the exception of structural currency exposures, all non-trading currency and commodity positions are subject to market risk regulatory capital and are therefore generally captured in VaR, although such positions do not contribute significantly to overall VaR.

Treasury also assumes market risk from its balance sheet and capital management responsibilities. Treasury finances non-monetary balance sheet items such as bank property and equity investments in associated companies; it also manages interest rate and foreign exchange risks resulting from the deployment of UBS's consolidated equity, from structural foreign exchange positions and from non-Swiss franc revenues and costs. The market risk limits allocated to Treasury cover both the risks resulting from these responsibilities, and those transferred from other business groups. The limits allow them flexibility to pre-hedge or delay hedging if desired, both to manage large flows and to take advantage of market movements.

Exposure to single names arising from debt instruments, such as loans, which are not originated or acquired as part of a trading activity is controlled under the credit risk framework. Neither idiosyncratic nor credit spread risk on these instruments is captured under the market risk framework. Credit spread risk is, however, a material component of the risk on the Investment Bank's syndicated financing business and the credit spread exposures from this activity are reported to senior management alongside those on the trading inventory.

The Investment Bank hedges an increasing proportion of its credit exposure. Specific hedges, such as credit default swaps on the same name, are reflected in credit risk measures, but other types of hedge may also be used for exposure management – for example, credit indices or proxy hedges on other names. Hedges of this type are treated as open positions for risk control purposes and are captured under the market risk framework.

Risk on equity investment positions, including private equity, is not controlled under the market risk framework.

Risk control organization, governance and structure

The Group Head of Market Risk, reporting to the Group Chief Risk Officer (Group CRO), is responsible for development of the market risk control framework. There is a CRO in each business group and a designated CRO for Treasury. The Group Head of Market Risk, the business group CROs and their teams are responsible for the independent control of market risk.

It is the primary responsibility of traders to identify the risks inherent in their activities, including those arising from new businesses and products, and from structured transactions. The independent controllers are responsible for ensuring that identified risks are completely and accurately captured in risk measurement systems and appropriately constrained by portfolio and concentration controls. They are also responsible for assessing the reasonableness of reported risk, particularly in relation to the revenues generated on the risk positions – an important step in identifying risks that are not adequately reflected in risk measures.

The Investment Bank CRO organization provides market risk measurement and reporting support to all business groups and, in close cooperation with the Group Head of Market Risk, is responsible for the development and ongoing enhancement of market risk measures, including the models used to measure VaR, stress loss and risk on tradable single name exposures.

The Chairman's Office delegates market risk authority to the GEB and approves delegations by the GEB ad personam to the Group CRO, the Group Head of Market Risk and the business group CROs. Further delegations are also made to market risk officers in the business groups. For many trading businesses, standard transactions within approved business lines and limits do not require prior risk control approval. Rather, risk management and risk control authority holders approve the retention of positions or give instructions for risk to be reduced based on subsequent review. Large transactions such as security underwritings and transactions creating less liquid risks – particularly structured and complex transactions – do, however, require pre-approval, as do temporary increases in limits to accommodate new transactions or positions.

Standard forms of market risk measures, limits and controls are applied to portfolios and risk concentrations. Other forms of measurement and control are developed, where necessary, for individual risk types, particular books and specific exposures. The quantitative controls are complemented by qualitative controls geared to the prompt identification, assessment, measurement and monitoring of market risks. Risks that are not well reflected by standard measures are subject to additional controls, potentially including transaction level pre-approval and specific limits.

UBS's policy requires that models used for valuation or which feed risk positions to risk control systems are subject to independent verification by specialist quantitative units in the CRO organization.

UBS's approach to valuation of instruments for which no directly observable market price is available is described in Note 26 in Financial Statements 2007. Valuation adjustments for model uncertainty are applied where appropriate, consistent with accounting requirements.

Reporting is an important component of the qualitative framework and UBS therefore has processes requiring regular reporting to senior managment in the business groups, the Group Head of Market Risk and Group CRO, and the GEB, Chairman's Office and Board of Directors (BoD).

Utilizations of most limits, including all major portfolio and concentration limits, are reported daily and all excesses are investigated. Daily reports, including commentary, on material risk positions are provided to senior management in the business groups and to the Group Head of Market Risk and Group CRO. Monthly and quarterly reports, including both quantitative and qualitative information, are prepared in the business groups, and these provide the basis for consolidated reports to the GEB, the Chairman's Office and the BoD.

Risk measures

UBS has two major portfolio measures of market risk – VaR and stress loss – which are common to all business groups. They are complemented by concentration risk measures and supplementary controls.

Concentration limits are tailored to the nature of the activities and the risks they create. They therefore differ between, for example, the Investment Bank, where the risks are most varied and complex, and Treasury, which carries market risk in a limited range of risk types and not generally in complex instruments.

Supplementary limits are established on portfolios, sub-portfolios, asset classes or products for specific purposes where standard limits are not considered to provide comprehensive control. These "operational limits" are intended to address concerns about, for example, market liquidity or operational capacity. They may also be applied to complex products for which not all model input parameters are observable, and which thus create difficulties in valuation and risk measurement. Operational limits can take a variety of forms including values (market, nominal or notional) or risk sensitivities. The ways in which operational limits are applied have been expanded following the experience of 2007.

Value at Risk (VaR)

VaR is a statistically based estimate of the potential loss on the current portfolio from adverse movements in both general and idiosyncratic market risk factors. The same VaR model is used for internal risk control (including limits) and as the basis for determining market risk regulatory capital requirements.

UBS measures VaR using a 10-day time horizon for internal risk measurement and control, and as the basis for market risk regulatory capital, and based on a 1-day horizon for information and for backtesting. VaR is derived from a distribution of potential losses. It expresses the amount that might be lost over the specified time horizon as a result of changes in market variables, but only to a certain level of confidence (99%) and there is therefore a specified statistical probability (1%) that actual loss over the period could be greater than the VaR estimate.

VaR is calculated at the end of each trading day, based on positions recorded at that time. Retrospective adjustments to valuations, affecting risk positions, may be booked some days later, particularly at period ends. VaR is not subsequently restated to reflect these later adjustments to positions.

VaR models are based on historical data and thus implicitly assume that market moves over the next 10 days or one day will follow a similar pattern to those that have occurred over 10-day and 1-day periods in the past. For general market risk, UBS uses a look-back period of five years – a period which generally captures the cyclical nature of financial markets and is likely to include periods of both high and low volatility. UBS applies these historical changes directly to current positions, a method known as historical simulation.

Idiosyncratic risk is measured on all forms of single name risk. For debt instruments the measure includes rating migration risk and prepayment risk – these measures were enhanced during 2007. For equity instruments, the measure is based on the Capital Asset Pricing Model ("CAPM") supplemented by a "deal break" methodology for equity arbitrage positions, where UBS is typically long in the stock of one company and short in that of another. The deal break measure assesses the probability of collapse of a merger or takeover, and its impact on the two stock prices – a one-off jump move generating the same potential loss for both 10-day and 1-day VaR.

The Chairman's Office annually approves a 10-day VaR limit for UBS as a whole and allocations to the business groups, the largest being to the Investment Bank. In the business groups, VaR limits are set for lower organizational levels as necessary.

In the start-up phase of a business, some market risks may be controlled outside VaR but the level of such risk is deliberately kept small, typically by application of operational limits.

Backtesting

The predictive power of the VaR model is monitored by "backtesting". Backtesting compares the 1-day VaR calculated on trading portfolios at close of each business day with the actual revenues arising on those positions on the next business day. These revenues ("backtesting revenues") exclude non-trading components such as commissions and fees, and estimated revenues from intraday trading. If backtesting revenues are negative and exceed the 1-day VaR, a "backtesting exception" is considered to have occurred. If the number of backtesting exceptions in a rolling 12-month period exceeds levels specified by regulators, the "multiplier" by which the market risk regulatory capital requirement is derived from 10-day VaR is increased.

Although UBS uses VaR to measure general market risks arising in non-trading books, the results are not formally backtested because these books are not generally marked to market.

VaR based on a 1-day horizon provides an estimate of the likely range of daily mark to market revenues on trading positions under normal market conditions. When 1-day VaR is measured at a 99% confidence level, such an exception can be expected, on average, one in a hundred business days. More frequent backtesting exceptions are likely to occur if market moves are greater than those seen in the look-back period, if the frequency of large moves increases, or if historical correlations and relationships between markets or variables break down (for example, in a period of market disruption or a stress event). Backtesting exceptions are also likely to arise if the way positions are represented in VaR does not adequately capture all their differentiating characteristics and the relationships between them. Such granularity can become particularly important as business grows and as markets evolve or when they experience the sort of dislocation seen in 2007.

UBS experienced many backtesting exceptions as a result of these developments and is responding to them.

All backtesting exceptions and any exceptional revenues on the profit side of the VaR distribution are investigated, and all backtesting results are reported to senior business management, the Group CRO and business group CROs.

As required by regulations, backtesting exceptions are also notified to internal and external auditors and relevant regulators.

Stress loss

The purpose of stress testing is to quantify exposure to extreme and unusual market movements. It is an essential complement to VaR.

The VaR measure is based on observed historical movements and correlations, whereas stress loss measures are informed but not constrained by past events. UBS's objectives in stress testing are to explore a wide range of possible outcomes, to understand vulnerabilities, and to provide a control framework that is comprehensive, transparent and responsive to changing market conditions.

UBS's stress scenarios include an industrial country market crash with a range of yield curve and credit spread behavior, and emerging market crises, with and without currency pegs breaking. A general recovery scenario is also assessed. The standard scenarios are run daily, and it is against these that the development of stress loss exposure is tracked and comparisons are made from one period to the next. Stress loss limits, approved by the Chairman's Office, are applied to the outcome of these scenarios for all business groups. Emerging markets stress loss in aggregate and stress loss for individual emerging market countries, measured under the standard stress scenarios, are also separately limited. The scenarios and their components are reviewed at least annually.

UBS has been developing and will soon implement a new approach to the specification of stress loss scenarios that better differentiates between the source of a stress event and its contagion effect.

Specific scenarios targeting current concerns and vulnerabilities are also used. They must, by definition, be constantly adapted to changing circumstances and portfolios. The choice of scenarios is judgmental, depending on management's view of potential economic and market developments and their relevance to the positions UBS carries. The market moves envisaged in a targeted stress test might prove to be less than the moves actually seen in a stress event, and actual events may differ significantly from those modeled in the stress scenarios. UBS's targeted stress tests did not predict the severe dislocation in US residential mortgage-related markets in 2007 – in particular the breakdown in correlation within and between asset classes and the complete drying up of liquidity. UBS is endeavoring to reflect these experiences in its stress testing framework.

The VaR results beyond the 99% confidence level are analyzed to better understand the potential risks of the portfolio and to help identify risk concentrations. The results of this analysis are valuable in their own right and can also be used to formulate position-centric stress tests. Although the standard scenarios incorporate generic elements of past market crises, more granular detail of specific historical events is provided by the VaR tail. During 2007, the "worst historical loss" from the VaR distribution was introduced as an additional formal stress scenario. UBS is also considering use of a longer historical time series, where available, to generate this stress exposure.

Additionally, UBS measures and limits the impact of increased default rates on the value of its portfolio of single name exposures.

UBS applies country ceilings to limit exposure to all but the best-rated countries and these measures cover market as well as credit risks.

Most major financial institutions employ stress tests, but their approaches differ widely and there is no benchmark or industry standard in terms of scenarios or the way they are applied to an institution's positions. Furthermore, the impact of a given stress scenario, even if measured in the same way across institutions, depends entirely on the make-up of each institution's portfolio, and a scenario highly applicable to one institution may have no relevance to another. Comparisons of stress results between institutions can therefore be highly misleading, and for this reason UBS, like most of its peers, does not publish quantitative stress results.

Concentration limits and other controls

UBS applies concentration limits to exposures to general market risk factors and to single name exposures. The limits take account of variations in price volatility and market depth and liquidity.

In the Investment Bank, limits are placed on exposure to individual risk factors. They are applied to general market risk factors or groups of highly correlated factors based on market moves broadly consistent with the basis of VaR, i.e. 10-day, 99% confidence moves. Each limit applies to exposures arising from all instrument types in all trading businesses of the Investment Bank. The market moves are updated in line with the VaR historical time series and the limits are reviewed annually or as necessary to reflect market conditions. The effectiveness of risk factor limits in controlling concentrations of risk depends critically upon the way risk positions are represented. If long and short positions are considered to be sensitive to the same risk factor, potential gains and losses from changes in that factor are netted. The steps UBS is taking to enhance granularity of risk representation in its VaR measure are equally relevant to its risk concentration controls.

The Investment Bank carries exposure to single names, and therefore to event – including default – risk. This risk is measured across all relevant instruments (debt and equity, in physical form and from forwards, options, default swaps and other derivatives including basket securities) as the aggregate change in value resulting from an event affecting a single name or group. The maximum amount that could be lost if all underlying debt and equity of each name became worthless is also tracked. Positions are controlled in the context of the liquidity of the market in which they are traded, and all material positions are monitored in light of changing market conditions and specific, publicly available information on individual names – market risk officers do not have access to non-public information and must therefore rely to a significant extent on external ratings.

This form of single name exposure measure is most appropriate to corporate clients, financial institutions and other entities, the value of whose equity and debt instruments is dependent on their own assets, liabilities and capital resources. Asset-backed securities are usually issued through special purpose, bankruptcy remote vehicles and it is more important to aggregate risk in other dimensions than the issuer name, in particular the factors affecting the value of the underlying asset pools. UBS monitored underlying exposures by broad asset category but not to the level of granularity that proved neccessary in the case of mortgage-backed securities. It is now enhancing its measures accordingly.

Exposures arising from security underwriting commitments are subject to the same measures and controls as secondary market positions. There are also governance processes for the commitments themselves, generally including review by a commitment committee with representation from both the business and the control functions. All firm underwriting commitments are approved under specific delegated risk management and risk control authorities.

Other applications of market risk measures

Market risk measurement tools may be selectively applied to portfolios for which the primary controls are in other forms. VaR can, for example, provide additional insight into the sensitivity of investment positions to market risk factors, even though some of the assumptions of VaR – in particular the relatively short time horizon – may not be representative of their full risk. The results can be used by business management and risk controllers for information or to trigger action or review.

Enhancements to market risk management and control

In 2007, UBS’s market risk measures underestimated the potential losses resulting from exposures to the previously deep and liquid US residential mortgage market – neither trading management nor market risk controllers foresaw the extreme rates of delinquency and default and low recovery levels now projected, or the breakdown in correlation within and between asset classes that emerged in the second half of 2007 and revealed the tail risk in UBS’s positions. With the accompanying drying up of liquidity in parts of the market, the size of UBS’s positions has proved excessive relative to the market.

The size, frequency and pattern of market moves were exceptional and UBS suffered losses in excess of those predicted by statistical or even stress loss risk measures based on historical market movements. Other market risk management and control processes were, however, intended to identify risk concentrations and sources of potential loss in more extreme circumstances. UBS is therefore taking steps to address the gaps which these events have revealed.

Risk management

Following the major losses, senior management changes were made. The fixed income, currencies and commodities (FICC) business unit is being restructured to build upon and strengthen its core business strategy of client servicing and risk distribution, and introduce stronger risk discipline:

– a workout group has been established to ensure robust risk management of the segregated, legacy portfolios and to develop orderly exit strategies;

– the remaining real estate-related activities are being refocused towards intermediation of client flows and alignment to the needs of investment banking and wealth management clients; and

– management of flow credit trading is being consolidated. This will improve risk aggregation and communication and, over time, will allow consolidation of risk management systems.

The introduction of a new funding framework for the Investment Bank will encourage more disciplined use of the balance sheet.

The models used in the risk management and valuation of US residential real estate-related products have been refined and recalibrated to reflect projections for lifetime cumulative losses consistent with market prices, where observable, and will continue to be updated as these projections and market parameters change. Even though some of these developments were undertaken under time pressure in a period of rapidly evolving markets and expectations, all models were approved for use by the independent model verification team. In first quarter 2008, the models were put into full production mode in the Investment Bank.

Risk control

As explained under “Risk measures”, which starts on page 33 of this section, where values of different instruments are assessed to be driven by the same risk factor, sensitivities in standard risk control measures have typically been expressed net across instruments and positions. If the drivers are not, in fact, the same risk factor but, rather, risk factors which have historically been very closely correlated, this netting will disguise “basis” risk – the risk of divergent movements between risk factors that are not perfectly correlated. A feature of the market dislocation in 2007 was the breakdown of historical correlations within markets, for example delinquency rates on sub-prime mortgages of different vintages. An important enhancement to the market risk control framework is therefore to increase the granularity of risk representation – not just for US residential mortgage related products but more generally – so that basis risks can be appropriately measured and controlled.

The specific characteristics of individual instruments which are critical in a stress event cannot always be predicted and it is therefore important to have a multi-faceted framework with complementary controls. UBS is applying more extensive limits, by asset class, based on gross values as well as risk sensitivities, in order to protect against extreme losses in the event of future dislocations and breakdowns – even if the probability of their occurring currently appears to be remote. Additionally, controls have been introduced to highlight positions which are large relative to market depth. In 2007, UBS also suffered significant markdowns as a result of leveraged positions – positions on which the rate of loss accelerated as market moves became more extreme – and on positions which were protected against market moves up to a certain point but fully exposed beyond that level. Such risks often only materialize when markets move beyond the range covered by statistical and stress loss parameters, and thus they do not show up in the regular risk measures. Controls over these deep tail risks already exist for some portfolios and are now being more widely applied. UBS is revising its approach to global stress testing to deliver a more diverse range of scenarios, which better differentiate between the source of a stress event and its contagion effect. Additionally, more extensive use of targeted stress tests is planned, using forward looking scenarios based on analysis of the positions and vulnerabilities of each portfolio. An important additional aspect of stress testing will be to consider liquidity as well as price sensitivity. New stress loss limits have been introduced at business area level, and on the stress loss contribution to standard scenarios from emerging markets, individually and in aggregate.

Since second quarter 2007, a series of in-depth reviews, which will cover all Investment Bank trading portfolios, has been under way. The findings of these reviews are being incorporated into development plans and will be factored into the quantitative enhancements described above. They also provide important insights into necessary qualitative changes to market risk management and control. It is intended that reviews of this type will become a regular feature, contributing to the ongoing enhancement of the market risk framework. UBS has, for many years, had a policy covering the pre-approval of structured and complex transactions. This has operated well at the transactional level but needs to be more systematically complemented by an assessment of the cumulative impact of “one-off” transactions with similar characteristhough tics on market risk in the portfolio. Processes have been adjusted so that repeat transactions now trigger a review as a new business initiative, the objective being to identify, at an early stage, any potential build-up of risks that are not appropriately captured by the control framework.

Investment Bank market risk policies are also being revised to emphasize the key characteristics of trading activities, and the criteria for assessing the liquidity of positions.

Market and credit risk control operate on opposite sides of information barriers – market risk officers do not generally have access to non-public information about clients and counterparties unless formally brought “over the wall”. There are, however, areas where the two teams can and do work closely together to provide a comprehensive and consistent view of risks. Areas for further cooperation and integration in organization and processes are being explored to better exploit the complementary skills of the two units. The planned quantitative improvements and enhancements to processes will be reinforced through education to ensure that the lessons from 2007 are understood by risk managers and risk controllers alike. UBS has a dedicated risk education unit, reporting to the Group CRO, which is developing programs to help strengthen risk culture and risk awareness.

Market risk in 2007

The graphs below illustrate the relative calm of the first half of 2007, and the severe dislocation of markets and extreme volatility of the second half. Markets retreated in response to the first fall in the US sub-prime market, but by the end of first quarter the impact appeared to have been contained. Equities markets resumed their upward trend from late March through to May, but by June inflation fears and renewed concerns about wider contagion from the sub-prime market led to increased volatility. A flight to quality and rapid deleveraging by some market participants followed in third quarter, resulting in intra-market dislocations as participants exited "crowded trades". Spreads between government yields and bank borrowing rates ("TED spread") widened dramatically and central banks intervened to counter the general drying up of liquidity. In the US residential mortgage-related markets, credit spreads on the highest rated securities reached unprecedented levels, with associated increases in volatilities. A temporary recovery in September was short-lived – markets fell again in November on further deterioration in US residential mortgage markets, fears of a US recession and uncertainties about the health of the banking sector. Despite a slight recovery in some markets in December many experienced professionals regard trading conditions in fourth quarter as amongst the most difficult for many years.

Value at Risk

In 2007, Investment Bank VaR was higher on average and at year-end than in 2006, with a much greater range between minimum and maximum.

The graph of US sub-prime mortgage-backed securities illustrates the scale of market value decline which eroded partial hedges on positions, and increased risk exposure. The extreme volatility of spreads has progressively increased reported VaR on existing positions with each update of the 5-year historical time series. While the illiquidity of large parts of this market has prevented significant reductions in positions, amortizations and writedowns, together with active risk reduction in other areas, resulted in a lower VaR at year end than the peak for the year which occurred in fourth quarter.

There were other notable contributors to the fluctuations in Investment Bank VaR during the course of the year. The acquisition of Banco Pactual, completed in December 2006, resulted in more volatile interest rate VaR. In the first half of the year, directional exposure to equity markets not only increased equity VaR but also reduced the offset against interest rate VaR. Credit spread exposure and idiosyncratic risk were major contributors to the risk profile throughout the year.

Market risk limits were adjusted in third quarter to constrain the level of risk in areas other than US residential mortgage market related exposure, but the impact on these positions of updates to the historical time series in third and fourth quarters has more than offset risk reductions elsewhere.

Corporate Center VaR was generally lower than in 2006, until fourth quarter when the major writedowns announced in December led to temporary positions associated with Treasury's management of the foreign exchange component of parent bank profits and losses, and its management of the parent bank equity.

VaR for UBS as a whole was dominated by the Investment Bank, with Treasury providing some offset at times.

Investment Bank: Value at Risk (10-day, 99% confidence, 5 years of historical data)

Year ended 31.12.07

Year ended 31.12.06

CHF million

Min.

Max.

Average

31.12.07

Min.

Max.

Average

31.12.06

Risk type

Equities

147

415

210

242

144

360

203

232

Interest rates (including credit spreads)

269

877

475

576

237

607

417

405

Foreign exchange

9

73

28

21

16

65

31

40

Energy, metals and commodities1

24

90

51

41

26

102

49

44

Diversification effect

2

2

(227)

(266)

2

2

(280)

(248)

Total

291

836

537

614

331

559

420

473

1 Includes base metals and soft commodities risk from 15 March 2006. 2 As the minimum and maximum occur on different days for different risk types, it is not meaningful to calculate a portfolio diversification effect.

UBS: Value at Risk (10-day, 99% confidence, 5 years of historical data)

Year ended 31.12.07

Year ended 31.12.06

CHF million

Min.

Max.

Average

31.12.07

Min.

Max.

Average

31.12.06

Business groups

Investment Bank1,2

291

836

537

614

331

559

420

473

Global Asset Management3

2

10

4

3

4

16

9

10

Global Wealth Management & Business Banking

2

5

3

3

4

14

10

5

Corporate Center

10

92

25

61

25

69

43

27

Diversification effect

4

4

(34)

(93)

4

4

(54)

(52)

Total

288

833

535

588

336

565

429

464

1 Includes UBS risk managed by Dillon Read Capital Management from June 2006 to 2 May 2007 and risk transferred from Dillon Read Outside Investor Fund from 3 May 2007. 2 Includes UBS Pactual from 1 December 2006. 3 Only covers UBS positions in alternative and quantitative investments. During first quarter 2007, seed money and coinvestments in these funds were reclassified as financial investments and they are not included in reported Value at Risk from that point. 4 As the minimum and maximum occur on different days for different business groups, it is not meaningful to calculate a portfolio diversification effect.

UBS: Value at Risk (1-day, 99% confidence, 5 years of historical data)1

Year ended 31.12.07

Year ended 31.12.06

CHF million

Min.

Max.

Average

31.12.07

Min.

Max.

Average

31.12.06

Investment Bank2

124

253

164

149

129

230

172

160

UBS

126

254

165

152

131

233

173

162

1 10-day and 1-day Value at Risk (VaR) results are separately calculated from underlying positions and historical market moves. They cannot be inferred from each other. 2 Positions in Investment Bank subject to market risk regulatory capital contributed average VaR of CHF 160 million in 2007 and CHF 169 million in 2006.

Backtesting

As a result of the severe market volatility and dislocation which prevailed from the end of July, UBS experienced 29 backtesting exceptions in 2007, its first since 1998. The multiplier by which the market risk regulatory capital requirement is derived from 10-day VaR has been increased accordingly.

The first histogram shows daily backtesting revenues alongside all daily revenues for 2007. In the second histogram, the daily backtesting revenues are compared with the corresponding VaR over the same 12-month period for days when backtesting revenues were negative.

Given market conditions, the occurrence of backtesting exceptions is not surprising. Moves in some key risk factors were, on occasions, well beyond the level expected statistically with 99% confidence based on the historical time series in use at the time. Despite regular updates to the time series, backtesting exceptions have continued, partly caused by "jump events". Some of these reflect a step change in market conditions, such as the mass downgrade by a rating agency of highly rated US residential mortgage market-linked securities. Others result from periodic new information or show the cumulative impact over several days or weeks of changing conditions in markets with diminished liquidity – the monthly remittance data published by mortgage servicers typically results in discontinuities of this type. UBS also made changes to its valuation approach to certain positions, leading to step changes. For example, as liquidity dried up, certain positions moved from a mark-to-market to a market-to-model basis. The enhancements UBS is making to market risk measures will not, of course, eliminate backtesting exceptions from these sources.

Stress loss

Stress loss for Investment Bank is defined as the worst case outcome from the official stress scenarios, which now include the worst historical loss from each day's VaR simulation. Stress loss, like VaR, is dominated by US residential mortgage-related positions, with a significant contribution from corporate credit spread exposure. Changing directional exposures to interest rates and equity markets have led to fluctuations in reported stress loss over the course of the year.

Value at Risk outlook

In its fourth quarter 2007 report, UBS indicated that it was considering changing the internal risk control and / or regulatory capital treatment of some US residential mortgage market-related positions. Decisions have now been taken to reclassify some of the legacy portfolios managed by the FICC workout group. The markets for these positions are not liquid and 10-day VaR is not an adequate measure of their risks or an appropriate risk control tool. They will no longer be subject to internal VaR limits, they will be subject to banking book, rather than trading book, regulatory capital, and they will be excluded from UBS's disclosed VaR from first quarter 2008.

The marginal contribution of the legacy positions to Investment Bank VaR at 31 December 2007 is approximately CHF 260 million.

Since year-end 2007, the VaR historical time series has been further updated to capture the continued increase in market volatility in November and December. As a result, VaR on existing Investment Bank positions – excluding the legacy positions – increased by approximately 4%. Had the legacy positions remained in VaR their marginal contribution to Investment Bank VaR would have more than trebled.

Pagina aggiornata il: 21 aprile 2008, 14.38

Audited information according to IFRS 7 and IAS 1
Risk disclosures provided in line with the requirements of the International Financial Reporting Standard 7 (IFRS 7), Financial Instruments: Disclosures, and disclosures on capital required by the International Accounting Standard 1 (IAS 1), Financial Statements: Presentation, form part of the financial statements audited by UBS’s independent registered public accounting firm Ernst & Young Ltd., Basel. This information (the audited texts, tables and graphs) is written in normal font throughout the report "Risk, Treasury and Capital Management 2007" and is incorporated by cross-reference into UBS’s Financial Statements 2007. Non-audited content is written in italic font.

¯¯¯¯

Important legal information - please read the disclaimer before proceeding.

Products and services in these webpages are not available for US persons, for the exclusion of residents of other nations see the disclaimers relating to the actual services.

© UBS 1998-2008. All rights reserved.

Privacy Policy