RETURN
AND RISK
GOVERNMENT
PENSION FUND
GLOBAL
/2016
No. 02
RETURN AND RISK GOVERNMENT PENSION FUND GLOBAL /2016 No. 02
Performance and risk 2015 Government Pension Fund Global
Contents 2016 1 2 3 INVESTMENTS RETURN RISK 1.1 The funds index________2 2.1 Return________________34 3.1 Risk___________________52 1.2 Investments___________28 2.2 Relative return________42 3.2 Relative risk___________56 2.3 Return and cost_______48 3.3 Return and risk________62 APPENDIX Fac
The funds index The total return on the fund over time will to a large extent be determined by developments in the broad markets in which the fund is invested. The strategic benchmark index defined in the management mandate from the Ministry of Finance captures developments in these markets. 2 Norge
INVESTMENTS against minimum standards of governance and investability. Eligible securities are assigned to a country and are required to pass screens for liquidity, free float and foreign ownership restrictions prior to being included. The benchmark index for equities deviates from the composition o
Table 1 The funds equity benchmark versus the FTSE Global All Cap Index (GEISAC) by country as at close of 31 December 2016 Deviation from FTSE Country Share of equity benchmark Percent Share of FTSE GEISAC index Percent UK 4 Percentage points Millions of kroner 198,280 10.3 6.1 4.2 Germany 5.3 2.9
INVESTMENTS Table 3 1.1 The funds fixed-income benchmark versus the Barclays Global Aggregate Index by sector as at close of 31 December 2016 Deviation from Barclays Share of fixed-income benchmark Percent Sector Inflation-linked bonds Share of Barclays Global Aggregate index Percent Percentage poin
THE FUNDS BENCHMARK INDEX HISTORY The first benchmark index for the fund was introduced in January 1998. The risk and return properties of the fund has been defined by the benchmark index, which thus can be seen as a representation of the investment strategy over time. Any deviations in the actual p
INVESTMENTS The currency distribution implied by the composition of the benchmark index has defined the currency basket in which overall benchmark index returns are measured. When the strategic benchmark index changes, the yardstick with which returns are measured changes as well. In the current ana
8 Investments Return and risk 2016 Government Pension Fund Global
INVESTMENTS THE ALLOCATION DECISION The equity share in the strategic benchmark index has been raised twice from a pure fixed income starting point. of the fund in 2007, and the decision was implemented from the end of June 2007. These equity share decisions were expected to dominate the risk return
THE EQUITY BENCHMARK INDEX The key discussions concerning the benchmark index for equity investments have been the composition of regions, markets and market segments. 10 Regional allocation The primary decision made on the equity benchmark index has been the regional allocation between three broad
INVESTMENTS Chart 2 Chart 2 The Regional Allocation of The Equity Index. Percent. The regional allocation of the strategic equity benchmark index compared to the market weighted regions like the FTSE World index (solid lines). Calculations are based on monthly constituent level data from FTSE. Perce
12 Investments Return and risk 2016 Government Pension Fund Global
INVESTMENTS Regional return Regional equity returns can deviate substantially for prolonged periods of time. Europe and Asia and Oceania outperformed Americas significantly in the 20032008 period, but their subsequent decline during the global financial crisis were more severe. After the financial c
Emerging market inclusion The benchmark index composition within the regions has changed over time. The initial 1998 equity benchmark index allocation was to 21 developed OECD countries equity markets. These markets were all part of the FTSE World index. The strategic direction has since been toward
INVESTMENTS 80 70 60 50 40 30 20 10 0 100 98 99 00 01 02 Index weight Europe 90 03 04 80 70 90 100 The regional allocation in both initial markets and markets later added to the strategic equity benchmark index. Calculations are based on monthly constituent level data from FTSE. Percent Chart 5 Sout
Smaller companies inclusion The FTSE World index, chosen for the inaugural 1998 benchmark index, cover large and medium-sized market capitalisation. In 2003, FTSE introduced its new FTSE All Cap, which included global representation of companies with a smaller capitalisation. In a letter of April 20
INVESTMENTS Chart 8 Chart 8 1 Capitalisation segments as share of the strategic equity benchmark index. Calculations are based on monthly constituent level data from FTSE. Percent 100 100 90 80 80 70 70 50 40 30 20 10 0 98 99 00 01 02 03 04 60 recommendation 90 05 Small capitalisation 06 07 08 60 50
Ethical exclusions The Ministry first issued guidelines for the observation and exclusion of companies from the Government Pension Fund Global in November 2004. The Ministry appointed a Council on Ethics to research and evaluate companies, and to make recommendations on exclusions based on the crite
INVESTMENTS Table 5 Return impact of equity benchmark index exclusions by year and exclusion criterion. Percentage points Criterion 2006 2007 2008 Product-based exclusions -0.01 -0.06 Production of specific weapon types -0.01 -0.06 2009 2010 2011 -0.04 0.11 -0.09 -0.04 0.17 0.03 -0.05 -0.12 Producti
THE FIXED-INCOME BENCHMARK INDEX The primary decision for the fixed-income benchmark index has been the regional allocation across the same broad geographical regions as the equity benchmark index. 20 Regional allocation When the Ministry implemented the inaugural 1998 benchmark index, the macroecon
INVESTMENTS 1 Chart 12 Chart 12 The regional allocation of the fixed-income benchmark index compared to global fixed-income markets as represented by Bloomberg Barclays Global Aggregate. Percent 100 90 90 70 60 50 40 30 20 10 0 98 80 70 60 50 40 America, 80 100 99 00 01 02 03 04 05 06 07 08 Index w
22 Investments Return and risk 2016 Government Pension Fund Global
INVESTMENTS The primary risk characteristic of the fixed- income benchmark index is the sensitivity of the market value to changes in interest rates. The analytical duration approximates this sensitivity. The difference in analytical duration between the benchmark index and the Global Aggregate mark
Emerging market bonds A broadening of the fixed-income benchmark index to include local currency debt from emerging market sovereigns has been assessed on several occasions. The Ministry announced their intention to expand the fixed-income benchmark index in the 2012 white paper on the management of
INVESTMENTS 1 Chart 15 Chart 15 The regional allocation of emerging markets as share of the fixed-income benchmark index. Percent 80 80 20 0 98 99 00 01 02 Index weight Europe 03 04 Index weight EM Asia and Oceania 05 06 07 08 09 Index weight EM Europe 10 11 Index weight America 60 40 60 100 Market
THE EFFECT OF ASSET CLASS BENCHMARK INDICES Substantial accumulated return differences have developed between the benchmark index of the fund and the broad publicly available alternatives. The broad indices are meant to be market proxies and the gains or losses relative to these proxies can be seen
INVESTMENTS 1 Chart 18 Return of the equity benchmark index relative to global equity markets as represented by FTSE World, or a combination of FTSE All World until September 2003 and FTSE All Cap afterwards. Return of the fixed-income benchmark relative to global fixed-income markets as represente
Investments The funds investments are diversified across asset classes, regions and sectors. The goal is to have well diversified investments that spread risk and generate high long-term return. 28 The fund is invested in three major asset classes, equities, fixed income and real estate. At the end
INVESTMENTS Table 6 Region Millions of kroner 1 North America 1 854 474 1 753 171 101 302 2.2 1 703 821 36.3 United States Canada Europe Sector composition of the funds equity holdings Sector Millions of kroner 1 Percent 39.5 Financials 1 093 314 23.3 37.4 Banks 502 511 10.7 Insurance 238 331 5.1 Fi
Table 8 30 1 Table 9 Currency composition of the funds bond holdings Currency Millions of kroner 1 Percent US Dollar 1 129 957 43.8 Euro 661 710 25.7 Japanese Yen 175 900 6.8 British Pound Sector composition of the funds bond holdings Sector Government bonds Government bonds Government-related bonds
INVESTMENTS 1.2 Table 10 Largest holdings of equities and bonds excluding sovereigns as at 31 December 2016. Covered bonds issued by financial institutions and debt issued by other underlying companies are included in the bonds. Millions of kroner Sector Nestlé SA Equities Bonds Total Consumer goo
Global investments NORTH AMERICA 32 EUROPE 2,268 companies 2,071 bonds from 582 issuers 400 properties 1 1,881 companies 1,584 bonds from 489 issuers 358 properties 1 LATIN AMERICA AFRICA 262 companies 173 bonds from 36 issuers 1 196 companies 16 bonds from 2 issuers Invest
INVESTMENTS MIDDLE EAST 152 companies 33 bonds from 12 issuers INTERNATIONAL ORGANISATIONS 128 bonds from 15 issuers 1 ASIA 3,898 companies 617 bonds from 78 issuers OCEANIA 328 companies 159 bonds from 36 issuers 33
Return The funds investment return was 6.92 percent in 2016 and has been 5.70 percent since inception. The funds total market value rose 35 billion kroner to 7,510 billion kroner in 2016. The investment return for the year was 447 billion kroner. However, the krone strengthened against the main curr
d: KEI 20170206 Table 11 Absolute return per year. Measured in the funds currency basket. Percent Chart 20 7 funds quarterly and and accumulated accumulated Chart The The funds quarterly annualised return. Percent Percent annualised return. Equity Fixed income Real estate 1 9.31 9.26 34.81 -0.99 12
36 Return Return and risk 2016 Government Pension Fund Global
RETURN 2.1 Table 12 Absolute return key figures. Measured in the funds currency basket. Annualised. Percent Since 01.01.1998 Last 10 years Last 5 years Last 3 years Return on equity investments 1 5.46 4.78 12.67 6.80 8.72 Return on fixed-income investments 4.84 4.37 3.62 3.81 4.32 Return on real es
38 BENCHMARK RETURN The funds equity benchmark returned 8.58 percent in 2016. Over the last three years the annualised investment return has been 6.73 percent. The second best equity sector in 2016 was basic materials at 24.77 percent. However, even after their recent gains, these two sectors still
RETURN 2.1 Table 15 Equity benchmark return by region and country. Annualised. Percent The funds currency basket Local currency 2016 3-Year 5-Year 2016 3-Year 5-Year 15.31 12.65 16.86 12.86 8.17 14.09 United States 14.65 13.27 17.73 12.42 8.25 14.54 Canada 27.85 3.68 5.93 21.02 7.08 8.89 1.79 1.97
Table 16 Equity benchmark return by sector. Annualised. Percent The funds currency basket 2016 3-Year 5-Year 2016 3-Year 5-Year Financials 8.30 6.05 14.05 8.47 6.39 14.32 Banks 10.03 3.08 11.45 10.39 4.41 12.43 Nonlife insurance 9.49 10.69 18.78 9.02 10.61 18.15 Life insurance 2.43 5.73 17.52 5.46
RETURN 2.1 Table 17 Fixed-income benchmark return by region and currency. Annualised. Percent The funds currency basket Local currency 2016 3-Year 5-Year 2016 3-Year 5-Year North America 5.48 7.26 4.85 3.15 3.12 2.44 US Dollar 5.38 7.82 5.28 3.32 3.04 2.43 Canadian Dollar 6.57 0.70 -0.21 0.88 4.01
Relative return The funds investment return was 15 basis points higher than the return on the funds benchmark in 2016 and has been 26 basis points higher since inception. 42 The investment returns on the funds equity and fixed-income investments can be compared with the returns on global benchmark i
d: KEI 20170209 Chart 22 The funds quarterly and accumulated Chart 11 The funds quarterly and accumulated annualised relative return. Percentage points annualised relative return. Percentage points 2.00 2.00 1.75 1.75 Table 19 Relative return per year. Measured in the funds currency basket. Percen
Table 20 Relative return. Measured in the funds currency basket. Annualised Since 01.01.1998 Last 10-years Last 5-years Last 3-years 2016 Return on equity and fixed-income investments (percent) 1 5.70 5.24 9.21 5.70 7.12 Return on equity and fixed-income benchmark (percent) 1 5.44 5.19 9.01 5.75 6.
RETURN INVESTMENT STRATEGIES The management of the fund is index close, but the contribution from all the investment strategies are a result of active management and decisions. Fund allocation decisions are made when rebalancing the benchmark portfolios exposure to a number of return drivers, the fu
Table 22 Contributions to relative return on equity and fixed-income investments from investment strategies in 2016. Percentage points Fund allocation Internal reference portfolio Fixed income Cross-asset allocation Total -0.04 -0.04 -0.02 -0.10 0.01 -0.05 0.00 -0.04 of which systematic factors 0.1
RETURN CONTRIBUTIONS PRIOR TO 2013 For the period 19992012, the relative return for the equity asset class can be decomposed into internal management and external management strategies. Internal management strategies for equities comprised asset management activities including transition, general ri
Return and costs Norges Bank maintains a high level of cost awareness in the management of the fund. Total management costs as a share of assets under management have been trending downwards for a number of years, despite the inclusion of additional markets, currencies and unlisted real estate. 48 T
RETURN 2.3 managers, as well as costs related to the internal team managing the external managers. Costs related to ownership strategies are allocated to internal security selection. Specific system costs are allocated to each strategy based on usage. Table 26 Management cost per investment strate
50 COST-ADJUSTED RELATIVE RETURN The funds relative return after management fees can be compared with the investment performance that could theoretically be expected to be achieved with a passive index management strategy. A passive investment strategy would aim at replicating a benchmark following
RETURN 2 Table 28 The funds relative return after management costs The fund's relative return before management costs The fund's management costs 3 years 5 years Since inception -5 20 26 -5 -6 -8 -10 14 17 3 years 5 years Since inception -3 -3 -5 5 6 6 Transaction costs related to replication of th
Risk The funds absolute risk is largely driven by its asset allocation. The expected absolute volatility of the fund was 10.6 percent at the end of 2016. 52 Market risk is defined as the risk of a decrease in the market value of the portfolio as a result of changes in financial market variables such
RISK Chart 27 Annual return 60 equity/40 fixed income. Chart 13 Annual return of 60 of equity/40 fixed income. Measured in dollars. Percent. Percent Measured in dollars. Chart 28 Annualised 5-year of 60 equity/40 Chart 14 Annualised 5-year rolling rolling return of return 60 equity/40 fixed income
EXPECTED ABSOLUTE VOLATILITY The funds expected absolute volatility, based on the statistical concept of standard deviation, shows how much the annual return on the funds investments can be expected to fluctuate and takes into account the correlation between different investments in the portfolio. V
RISK BREAKDOWN OF EXPECTED ABSOLUTE VOLATILITY The expected volatility of equity investments was 14.0 percent at the end of 2016. A decomposition of the portfolio by industry shows that investments in financials contributed the most to the volatility in the portfolio, with 3.6 percentage points. Thi
Relative risk Deviations from the benchmark are sources of relative risk. This section looks at different approaches to measuring relative risk in the fund. 56 The composition of the fund differs from its benchmark indices along several dimensions including currencies, sectors, countries, regions, i
RISK EXPECTED ABSOLUTE AND RELATIVE VOLATILITY Expected absolute volatility estimates how much the annual return on the funds investments can be expected to fluctuate, while expected relative volatility, or expected tracking error, estimates how much the annual return on the funds equity and fixed-i
Table 33 Relative risk contribution to equity investments as at 31 December 2016. Basis points Sector Financials 58 Relative volatility contribution 11 Table 34 Relative risk contribution to fixed-income investments as at 31 December 2016. Basis points Relative volatility contribution Sector Gove
RISK EXPECTED SHORTFALL Expected relative volatility is an estimate of what happens under normal market conditions, but provides no information about the distribution and magnitudes of less probable outcomes (tail risk). Expected shortfall, also called conditional value at risk, is widely used as a
60 BENCHMARK OVERLAP Benchmark overlap is an important relative risk measure part of relative risk and shows how closely the portfolios match the benchmark indices. In line with the management mandate from the Ministry of Finance, Norges Banks Executive Board has set a limit for minimum overlap betw
RISK 3.2 Table 38 Characteristics of the distribution for realised monthly relative return. Measured in the funds currency basket Since 1998 1 Last 10-years Last 5-years Last 3-years Standard deviation relative return of equity and fixed-income invest- ments (percent) 0.20 0.26 0.11 0.11 Skewness r
Return and risk This section looks at various risk-adjusted performance measures and factor-adjusted regression analysis of returns. 62 The returns discussed in the previous sections of this report are useful for assessing the funds achievements against its long-term targets. However, it is not appr
RISK Sharpe ratio The Sharpe ratio is a widely used risk-adjusted performance measure. The Sharpe ratio is computed by dividing the average portfolio return in excess of the risk-free rate by the standard deviation of portfolio returns. A higher Sharpe ratio indicates a higher expected reward per un
Although fixed-income investments have often had lower average returns than equity investments, the returns have also been less volatile resulting in higher Sharpe ratios in periods such as 20082012, which includes the financial crisis. Comparing fixed-income investments with the benchmark, the rela
RISK Jensens alpha Under the assumptions of the Capital Asset Pricing Model (CAPM), all differences in expected return are explained by beta. Beta measures systematic risk and is estimated using a regression of the portfolio returns in excess of the risk-free rate on the benchmark excess returns. Je
66 FACTOR-ADJUSTED RETURN The analyses introduced here involve multivariate regressions of relative returns against sets of historical factor return series. Estimated regression coefficients are interpreted as active exposures to systematic factors over the historical period. Regression intercepts c
RISK 3.3 Table 46 Equity investments. Regression analysis of relative return in dollars before management costs Profitable expensive (HML) Sample period Regression coefficients Since 01.01.1999 38 0.02 0.05 -0.01 0.01 -0.02 45 Last 10 years 32 0.02 0.03 0.00 -0.01 -0.04 45 Last 5 years 27 0.01 0.02
68 Appendix Return and risk 2016 Government Pension Fund Global
APPENDIX 4 Appendix 69
Contents 1 Factor-adjusted returns 1.1 Introduction . . . . . . 1.2 Data and methodology 1.3 Results . . . . . . . . . 1.4 Factor return statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDIX 42 43 44 4 Appraisal ratio before management costs for moving sample periods . . . . . . . . 110 Appraisal ratio after management costs for various sample sizes . . . . . . . . . . . 110 Appraisal ratio after management costs for moving sample periods . . . . . . . . . 110 List of Figures 1
1 Factor-adjusted returns 1.1 Introduction This part of the appendix aims to shed light on the robustness of the estimated alphas and factor exposures reported in the Return and risk section. We present results from several factor regressions using alternative model specifications and differ- ent sa
APPENDIX 4 analysis of the results. For the fixed income factors, relevant data is sourced from Barclays in order to construct the factor returns. Table 3 provides a summary of all the factor abbreviations and data sources used. The data sourced from Fama and French, AQR and Barclays were downloaded
MKT, SMB, HML and UMD should resemble the equivalents found in the global factor returns from Fama and French (1992) and Carhart (1997), but due to minor differences in sorting pro- cedures and country neutralisations, discrepancies are expected. Correlations between the factors can be found in Sect
APPENDIX 4 Default premium factor (DEF) The default premium is defined in Fama and French (1993) as the difference between returns of corporate bonds and treasury bonds with more than 10 years to maturity. Table 2 summarises the sources that have been used to create the default premium factor return
Table 3 Summary information about the factors used for regressions Factor Description MKT Equity market return in excess of the risk free rate F-F AQR Jan 1998 to Dec 2016 Jan 1998 to Dec 2016 SMB Small Minus Big, return spread between small cap and large cap stocks F-F AQR Jan 1998 to Dec 2016 Jan
APPENDIX 4 1.3 Results In the next sections, separate regression results for the funds equity and fixed-income investments are presented, followed by regression results for the equity and fixed-income investments combined. Regression results are shown using relative returns after management costs, c
Table 5 Equity five-factor size-constrained regressions Full-period regression results with Fama-French global return factors. The dependent variable is the monthly return on the equity portfolio relative to the equity benchmark after management costs. All of the models are based on the 5-factor Fam
APPENDIX 4 Table 6 presents the average equity relative return after management costs, along with estimated results from four different factor model specifications typically employed in empirical asset pricing. The table provides insights into the sensitivity of the results starting from a one-facto
Table 7 Equity three- and four-factor size-constrained regressions Full-period regression results with Fama-French global return factors. The dependent variable is the monthly return on the equity portfolio relative to the equity benchmark after management costs. Model (1) is the 3-factor Fama-Frenc
APPENDIX 4 Tables 8 to 10 present regression results for the three-, four- and five-factor models for selected time periods, using the original factor returns and size-constrained factor returns. These tables give some indication of the sensitivity of the estimated parameters to varying time periods
Table 9 Equity four-factor size-constrained regressions for selected time periods Regression results with Fama-French global return factors for selected time periods. The dependent variable is the monthly return on the equity portfolio relative to the equity benchmark after management costs. Model (
APPENDIX 4 Table 10 Equity five-factor size-constrained regressions for selected time periods Regression results with Fama-French global return factors for selected time periods. The dependent variable is the monthly return on the equity portfolio relative to the equity benchmark after management co
Table 11 compares regression results for the five-factor model using equity relative returns since inception before and after management costs. Table 11 Equity five-factor size-constrained regressions before and after management costs Regression results before and after management costs with the 5-f
APPENDIX 4 Table 12 Equity three-, four- and six-factor regressions using AQR return series Full-period regression results with AQR global return factors. The dependent variable is the monthly return on the equity portfolio relative to the equity benchmark after management costs. Model (1) is a 3-fa
Table 13 shows results for the full six-factor model for different periods. The choice of HML factor affects the estimated sensitivity to the UMD factor (momentum) for the entire sample period and over the last 10 years, but has a limited impact on the momentum factor estimated using the last 5 year
APPENDIX 4 Table 14 Fixed-income two-factor regressions for selected time periods Regression results with global fixed-income factor returns constructed from Barclays data for selected time periods. The dependent variable is the monthly return on the fixed-income portfolio relative to the fixed-inco
The global fixed-income factors have different currency compositions between long term treasuries and short term treasuries. Table 16 reports regression results for the same specifications as in Table 15, but with the factors consisting only of US dollar denominated bonds. Changes are minor for most
APPENDIX 4 Equity and fixed-income investments Table 18 presents the regression results for the relative return on equity and fixed-income invest- ments combined after management costs using the seven-factor model recommended in Dahlquist et al. (2015) for different sample periods with the original
Table 19 presents regression results using original and size-constrained factor returns. The alpha estimates are close to zero for both the model incorporating only original factors and the model incorporating all of the size-constrained factors. The funds equity and fixed-income investments combine
APPENDIX 4 Table 20 presents the regression results for the combined equity and fixed-income relative returns after management costs with the seven-factor model for different sample periods using the size- constrained Fama-French factor returns. Table 20 Fund (ex-real estate) size-constrained factor
Table 21 presents the regression results since inception before and after costs using original and adjusted factors. Table 21 Fund (ex-real estate) factor regressions before and after management costs Full period regression results with global 7-factor model. Factor return series are based on Fama-F
APPENDIX 4 1.4 Factor return statistics To inform the interpretation of the previous results, we now present some statistics on the factors used in this appendix. We show statistics relating to factor returns, time-series characteristics and correlations between factors. Figure 1 shows the cumulativ
The global Fama-French factors are a simple average of the factors constructed in small cap stocks and large-cap stocks. As seen in Figures 2 through 5 the cumulative return is different for small-cap and large-cap stocks for the value, momentum, profitability and investment factors. Figure 2 Cumula
APPENDIX 4 Figure 4 Cumulative returns, global RMW factor along with Big and Small versions, 1998-2016 400 300 200 100 0 2000 2005 FF RMW FF RMW Big 2010 2015 FF RMW Small Figure 5 Cumulative returns, global CMA factor along with Big and Small versions, 1998-2016 400 300 200 100 0 2000 2005 FF CMA F
Figure 6 shows the cumulative returns of the factors from AQR Capital Management. Figure 6 Cumulative returns, global AQR factors, 1998-2016 500 400 300 200 100 0 2000 96 2005 2010 AQR MKT AQR HML lag AQR UMD AQR SMB AQR HML cur AQR QMJ 2015 AQR BAB Figure 7 shows the cumulative return of the fixed-
APPENDIX 4 Table 22 Factor return statistics since 1998 Arithmetic average return and volatility of monthly returns (annualised) over the period 1998-2016 for all factors. Figures are annualised with simple distributional assumptions of independence and stationarity. Factor Average return Volatility
Table 23 Factor return statistics for the last 10 years Arithmetic average return and volatility of monthly returns (annualised) over the period 2006-2016 for all factors. Figures are annualised with simple distributional assumptions of independence and stationarity. Factor Average return Volatility
APPENDIX 4 Table 24 Factor return statistics for the last 5 years Arithmetic average return and volatility of monthly returns (annualised) over the period 2011-2016 for all factors. Figures are annualised with simple distributional assumptions of independence and stationarity. Factor Average return
Tables 25 to 28 show the linear correlation between monthly factor returns. Table 25 Correlations between the Fama-French-Carhart factors and fixed income factors F-F MKT F-F SMB F-F HML F-F WML F-F RMW F-F CMA DEF Adj TERM 1.00 -0.01 -0.15 -0.27 -0.47 -0.46 0.47 -0.13 1.00 0.03 0.20 -0.22 -0.04 0.0
APPENDIX 4 return. In principle, this is not consistent with measuring the fund and benchmark returns in the currency basket. On the other hand, there is no established alternative. 2.1 Methodology In the following section, the methods used for calculating risk-adjusted measures and confidence inter
where μ rrel is the sample average of relative returns, and σ rrel is the sample standard deviation of relative returns using the T 1 divisor. The annualised information ratios and the corresponding confidence intervals are computed in the same way as for the Sharpe ratio. Jensens alpha The Capital
APPENDIX 4 2.2 Results In this section, 95 percent confidence intervals for all the risk-adjusted measures are reported before and after management costs. Results are computed since inception, for the last 10 years, the last 5 years and for 5-year rolling windows. Sharpe ratio Tables 29 to 32 report
Table 30 Sharpe ratio before management costs for moving sample periods Annualised Sharpe ratio estimates before costs for moving sample periods, along with 95 percent confi- dence intervals (parentheses). The estimates are based on monthly returns of equity, fixed-income and combined portfolios and
APPENDIX 4 Table 32 Sharpe ratio after management costs for moving sample periods Annualised Sharpe ratio estimates after costs for moving sample periods, along with 95 percent confidence intervals (parentheses). The estimates are based on monthly returns of equity, fixed-income and combined portfol
Information ratio Tables 33 through 36 report information ratios along with confidence intervals before and after management costs. Table 33 Information ratio before management costs for various sample sizes Annualised information ratio estimates before costs for various sample periods, along with 9
APPENDIX 4 Table 36 Information ratio after management costs for moving sample periods Annualised information ratio estimates after costs for moving sample periods, along with 95 percent confidence intervals (parentheses). The estimates are based on monthly returns of equity, fixed-income and combin
Table 38 Jensens alpha before management costs for moving sample periods Annualised Jensens alpha estimates before costs (percent) for moving sample periods, along with 95 percent confidence intervals (parentheses) and the R-squared from a regression of relative return on a constant and the benchmar
APPENDIX 4 Table 40 Jensens alpha after management costs for moving sample periods Annualised Jensens alpha estimates after costs (percent) for moving sample periods, along with 95 percent confidence intervals (parentheses) and the R-squared from a regression of relative return on a constant and the
Table 42 Appraisal ratio before management costs for moving sample periods Annualised appraisal ratio estimates before costs for moving sample periods, along with 95 percent confidence intervals (parentheses). The estimates are based on monthly returns on the equity, fixed- income and combined portf
APPENDIX 4 References Asness, Clifford S. and Frazzini, Andrea. The Devil in HMLs Details. Available at SSRN: http://ssrn.com/abstract=2054749, 2011. Asness, Clifford S., Frazzini, Andrea, and Pedersen, Lasse Heje. Quality Minus Junk. Available at SSRN: http://ssrn.com/abstract=2312432, 2014. Asvanu
Treynor, Jack L. Toward a Theory of Market Value of Risk Assets. Unpublished manuscript, 1962. Treynor, Jack L. and Black, Fischer. How to Use Security Analysis to Improve Portfolio Selection. Journal of Business vol. 46(1), pp. 66-85, 1973. 112 Appendix Return and risk 2016 Government Pension Fun
APPENDIX 4 113
Table 49 Risk-adjusted measures for equity and fixed-income investments. Before and after management costs. Annualised Since 01.01.1998 Last 10 years Last 5 years Last 3 years Before costs After costs Before costs After costs Before costs After costs Before costs After costs Mean return equity and
APPENDIX 4 Table 50 Risk-adjusted measures for equity and fixed-income investments. Before and after management costs. Annualised 19982002 20032007 20082012 20132016 Before costs After costs Before costs After costs Before costs After costs Before costs After costs Mean return equity and fixed-inco
Table 51 Risk-adjusted measures for equity investments. Before and after management costs. Annualised Since 01.01.1999 Last 10 years Before costs After costs Before costs Mean return equity investments (percent) 6.41 6.28 Mean return equity benchmark index (percent) 5.89 5.89 Mean relative return e
APPENDIX 4 Table 52 Risk-adjusted measures for equity investments. Before and after management costs. Annualised 19992002 Before costs After costs 20032007 Before costs 20082012 After costs Before costs After costs 20132016 Before costs After costs Returns Mean return equity investments (percent) -
Table 53 Risk-adjusted measures for fixed-income investments. Before and after management costs. Annualised Since 01.01.1998 Before costs Last 10 years Last 5 years After costs Before costs After costs Before costs After costs Last 3 years Before costs After costs Returns Mean return fixed-income i
APPENDIX 4 Table 54 Risk-adjusted measures for fixed-income investments. Before and after management costs. Annualised 19982002 20032007 20082012 Before costs After costs Before costs After costs Before costs Mean return fixed-income investments (percent) 6.13 6.08 3.98 3.92 5.81 Mean return fixed-
ISSN 2535-3225 Design: Brandlab Photos: Norges Bank Investment Management, Offset, Shutterstock Paper: Galerie Art Silk 300 g / Matt 150 g Production: 07 Media AS Print run: 350
NORGES BANK INVESTMENT MANAGEMENT Bankplassen 2, P.O. Box 1179 Sentrum, NO-0107 Oslo, Norway T: +47 24 07 30 00, www.nbim.no