Deutsche Bank

Annual Report 2017

Trading Market Risk Exposures

Value-at-Risk Metrics of Trading Units of Deutsche Bank Group (excluding Postbank)

The tables and graph below present the value-at-risk metrics calculated with a 99 % confidence level and a one-day holding period for our trading units. They exclude contributions from Postbank trading book which are calculated on a stand-alone basis.

Value-at-Risk of our Trading Units by Risk Type

 

Total

Diversification effect

Interest rate risk

Credit spread risk

Equity price risk

Foreign exchange risk1

Commodity price risk

in € m.

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

1

Includes value-at-risk from gold and other precious metal positions.

Average

29.8

32.0

(28.1)

(35.0)

20.2

19.7

19.7

26.6

8.7

9.3

8.4

10.7

0.8

0.7

Maximum

38.4

59.4

(37.6)

(57.6)

26.0

29.5

25.1

32.5

12.5

52.4

16.5

16.7

3.0

3.3

Minimum

20.1

20.4

(21.4)

(25.6)

13.5

14.8

13.5

22.3

4.4

4.4

4.2

3.6

0.1

0.2

Period-end

29.1

30.1

(22.5)

(36.9)

21.4

19.9

14.4

24.3

10.1

10.0

4.9

12.6

0.7

0.2

Development of value-at-risk by risk types in 2017

The average value-at-risk over 2017 was € 29.8 million, which is a decrease of € 2.2 million compared with the full year 2016. The average credit spread value-at-risk decreased due to a reduction in idiosyncratic risk.

The period end value-at-risk reduction was driven by reductions across the credit spread and foreign exchange asset classes.

Regulatory Trading Market Risk Measures (excluding Postbank)

The table below presents the stressed value-at-risk metrics calculated with a 99 % confidence level and a one-day holding period for our trading units. It excludes contributions from Postbank’s trading book which are calculated on a stand-alone basis

Average, Maximum and Minimum Stressed Value-at-Risk by Risk

 

Total

Diversification effect

Interest rate risk

Credit spread risk

Equity price risk

Foreign exchange risk1

Commodity price risk

in € m.

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

1

Includes value-at-risk from gold and other precious metal positions.

Average

76.7

85.2

(88.4)

(78.2)

69.8

51.9

62.1

74.9

18.8

20.6

12.6

14.8

1.8

1.3

Maximum

125.0

143.7

(115.8)

(150.0)

92.0

82.5

73.2

99.3

66.8

144.5

28.0

30.4

6.1

3.9

Minimum

42.0

60.4

(73.0)

(53.4)

48.3

37.4

54.3

59.0

1.5

2.4

6.9

3.4

0.3

0.4

Period-end

85.6

75.8

(81.0)

(91.3)

67.8

51.9

64.3

63.0

19.9

29.6

12.6

22.1

1.9

0.5

The average stressed value-at-risk was € 76.7 million over 2017, a decrease of € 8.5 million compared with the full year 2016. The reduction in the average was driven by a decrease in credit spread stressed value-at-risk due to a reduction in idiosyncratic risk as well as a small reduction coming from a model enhancement to the credit spread component. This has been partly offset by an increase in interest rate stressed value-at-risk due to a change in directional exposure on average over 2017.

The following graph compares the development of the daily value-at-risk with the daily stressed value-at-risk and their 60 day averages, calculated with a 99 % confidence level and a one-day holding period for our trading units. Amounts are shown in millions of euro and exclude contributions from Postbank’s trading book which are calculated on a stand-alone basis.

Development of value-at-risk and stressed value-at-risk in 2017

For regulatory reporting purposes, the incremental risk charge for the respective reporting dates represents the higher of the spot value at the reporting dates, and their preceding 12-week average calculation.

Average, Maximum and Minimum Incremental Risk Charge of Trading Units (with a 99.9 % confidence level and one-year capital horizon)1,2,3

 

Total

Non-Core Operations Unit

Global Credit Trading

Core Rates

Fixed Income & Currencies APAC

Emerging Markets – Debt

Other

in € m.

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

1

Amounts show the bands within which the values fluctuated during the 12-weeks preceding December 31, 2017 and December 31, 2016, respectively.

2

Business line breakdowns have been updated for 2017 reporting to better reflect the current business structure.

3

All liquidity horizons are set to 12 months.

Average

802.1

840.2

0.0

52.0

544.6

393.0

107.1

200.4

168.1

188.6

37.2

116.8

(54.8)

(110.5)

Maximum

899.3

944.4

0.0

57.3

597.4

405.8

172.5

229.6

229.0

243.0

62.9

128.0

(20.4)

(65.6)

Minimum

754.8

693.0

0.0

44.5

503.7

368.0

48.7

173.7

92.4

119.6

(1.4)

111.6

(90.0)

(141.8)

Period-end

789.6

693.0

0.0

51.8

540.1

368.0

133.2

173.7

142.3

119.6

19.9

121.8

(45.9)

(141.8)

The incremental risk charge as at the end of 2017 was € 790 million an increase of € 97 million (14 %) compared with year end 2016. The 12-week average of the incremental risk charge as at the end of 2017 was € 802 million and thus € 38 million (5 %) lower compared with the average for the 12-week period ended December 31, 2016. The decreased average incremental risk charge is driven by a decrease in credit exposures in the Core Rates and Emerging Markets Debt business areas when compared to the full year 2016.

For regulatory reporting purposes, the comprehensive risk measure for the respective reporting dates represents the higher of the internal spot value at the reporting dates, their preceding 12-week average calculation, and the floor, where the floor is equal to 8 % of the equivalent capital charge under the standardized approach securitization framework.

Average, Maximum and Minimum Comprehensive Risk Measure of Trading Units (with a 99.9 % confidence level and one-year capital horizon)1,2,3

in € m.

2017

2016

1

Regulatory Comprehensive Risk Measure calculated for the 12-week period ending December 29.

2

Period end is based on the internal model spot value.

3

All liquidity horizons are set to 12 months.

Average

5.4

31.3

Maximum

6.3

39.8

Minimum

4.5

21.9

Period-end

4.4

17.9

The internal model comprehensive risk measure as at the end 2017 was € 4.4 million a decrease of € 13.5 million (−75 %) compared with year end 2016. The 12-week average of our regulatory comprehensive risk measure as at the end of 2017 was € 5.4 million and thus € 25.8 million (83 %) lower compared with the average for the 12-week period ending December 31, 2016. The reduction was due to continued de-risking on this portfolio.

Market Risk Standardized Approach

As of December 31, 2017, the securitization positions, for which the specific interest rate risk is calculated using the market risk standardized approach, generated capital requirements of € 379.5 million corresponding to risk weighted-assets of € 4.7 billion. As of December 31, 2016 these positions generated capital requirements of € 278.4 million corresponding to risk weighted-assets of € 3.5 billion.

For nth-to-default credit default swaps the capital requirement decreased to € 2.8 million corresponding to risk weighted-assets of € 35 million compared with € 6.4 million and € 80 million as of December 31, 2016.

The capital requirement for Collective Investment Undertakings under the market risk standardized approach was € 45 million corresponding to risk weighted-assets of € 556 million as of December 31, 2017, compared with € 39 million and € 487 million as of December 31, 2016.

The capital requirement for longevity risk under the market risk standardized approach was € 32 million corresponding to risk-weighted assets of € 395 million as of December 31, 2017, compared with € 46 million and € 570 million as of December 31, 2016.

Value-at-Risk at Postbank

The value-at-risk of Postbank’s trading book calculated with a 99 % confidence level and a one-day holding period amounted to zero as of December 31, 2017. Postbank’s current trading strategy does not allow any new trading activities with regard to the trading book. Therefore, Postbank’s trading book did not contain any positions as of December 31, 2017. Nevertheless, Postbank will remain classified as a trading book institution.

Results of Regulatory Backtesting of Trading Market Risk

In 2017 we observed three global outliers, where our loss on a buy-and-hold basis exceeded the value-at-risk of our Trading Books, compared with one outlier in 2016. The outliers in 2017 all occurred in the fourth quarter. The first was driven by an idiosyncratic event that led to losses in our Non-Strategic and Emerging Markets Debt business areas. The second was an idiosyncratic event primarily impacting our Equities business. The final outlier was at the year end and was caused by losses across a number of business areas. The first and third of these events also led to an outlier on an Actual Backtesting basis, which compares the VaR to Total Income less Fees & Commissions, and excluding Debt Valuation Adjustments. There were two Actual Backtesting outliers in 2017 compared to four in 2016.

Based on the backtesting results, our analysis of the underlying reasons for outliers and enhancements included in our value-at-risk methodology, we continue to believe that our value-at-risk model will remain an appropriate measure for our trading market risk under normal market conditions. The following graph presents trading units’ daily comparison of the VAR measure as of the close of the previous business day with both hypothetical (buy-and-hold income, i.e. one-day change in portfolio’s value) and the actual backtesting outcomes (as defined above), in order to highlight the frequency and the extent of the backtesting exceptions. The value-at-risk is presented in negative amounts to visually compare the estimated potential loss of our trading positions with the buy and hold income. The chart shows that our trading units achieved a positive buy and hold income for 57 % of the trading days in 2017 (versus 54 % in 2016), as well as displaying the global outliers experienced in 2017.

The capital requirements for the value-at-risk model, for which the backtesting results are shown here, accounts for 1.3 % of the total capital requirement for the Group.

EU MR4 – Comparison of VAR estimates with gains/losses

Daily Income of our Trading Units

The following histogram shows the distribution of daily income of our trading units (excluding Postbank). Daily income is defined as total income which consists of new trades, fees & commissions, buy & hold income, reserves, carry and other income. It displays the number of trading days on which we reached each level of trading income shown on the horizontal axis in millions of euro.

Distribution of daily income of our trading units in 2017

Our trading units achieved a positive revenue for 93 % of the trading days in 2017 compared with 87 % in the full year 2016.