2 Management performance
2.1 Overview
The value of the Government Pension Fund was NOK 1,890.6 billion as per yearend 2006. The NOK 309.1 billion increase from the previous year was caused by the inflow of new funds, very favourable returns in the fixed-income markets, relatively low returns in the bond markets and excess returns from operational management. A certain appreciation of the Norwegian krone, as measured against the currency basket of the Government Pension Fund – Global, contributed negatively, but the Norwegian krone exchange rate is not relevant as far as developments in terms of international purchasing power are concerned. In Figure 2.1, the increase in value during 2006 is decomposed across various factors.
Figure 2.1 Development in the market value of the Government Pension Fund during 2006, as attributed to various components. NOK billion.1
1 In line with the resolution passed by the Storting in connection with the balancing of the Fiscal Budget for 2006, NOK 111.8 billion of the fund capital (measured by market value) were written down on 29 December 2006, as part of the termination of the arrangement whereby fund liquidity was held in the form of sight deposits with the treasury. At the same time, NOK 10 billion of the funds held in sight deposits were repaid, thus enabling a strengthening, through the Government Pension Fund – Norway, of long-term state ownership in the Norwegian business sector in line with the Soria Moria Declaration. The transaction is described in more detail in Proposition No. 31 (2006-2007) to the Storting, New balancing of the Fiscal Budget for 2006.
Source Ministry of Finance
Figure 2.2 shows developments in the market value of the Government Pension Fund since 1996. The value of the Government Pension Fund – Global and the Government Pension Fund – Norway was NOK 1,783.7 and 106.9 billion, respectively, as per yearend 2006. 1 This represents a NOK 393.6 billion increase since yearend 2005 as far as the Global part of the Fund is concerned, whilst the value of the Government Pension Fund – Norway as per yearend 2006 was less than at the beginning of the year as the result of the termination of the sight deposits with the treasury. NOK 288.3 billion of the increase in the value of the Government Pension Fund – Global is caused by the inflow of new capital from the Fiscal Budget. Total inflow to the Pension Fund – Global over the years 1996–2006 was NOK 1,440.6 billion. 2
Figure 2.2 The market value of the Government Pension Fund. 1996-2006.1
1 The Government Pension Fund was created in 2006 as a general framework encompassing the Petroleum Fund and the National Insurance Scheme Fund. The value of the two Funds has, for purposes of illustration, been aggregated for previous years as well.
Source Ministry of Finance
The highly favourable development in the value of the Government Pension Fund during 2006 is influenced by a number of factors. The significant easing of monetary policy in many countries during the period 2000–2003 has, together with the increased participation of countries like China and India in world trade, stimulated economic growth, and enterprises have registered high earnings. At the same time, inflation has remained low, cf. Figure 2.3a. This has paved the way for a broad stock market upturn, cf. Figure 2.3b. The upturn on the Oslo Stock Exchange in recent years has been particularly strong, and needs to be seen in the context of, inter alia , higher oil prices. However, in the last couple of years there has been a gradual tightening of US and European monetary policy, and bond yields have increased, cf. Figure 2.3c.
Figure 2.3 Market developments
Source EcoWin, Norges Bank and www.econ.yale.edu/shiller
The expected excess return on equities relative to risk-free investments (the “equity premium”) now appears, in line with the assumptions outlined in Chapter 3, to be relatively low from a historical perspective. Figure 2.3d shows that the ratio between price and earnings in the US stock market remains at a relatively high level from a historical perspective (adjusted for inflation and smoothed across business cycles). This may indicate a relatively low expected equity premium, although the current interest rate level and expected earnings growth also contribute to increasing the ratio between price and earnings. Returns in the global stock market in recent years have been significantly higher than the long-term estimates for return on equities adopted in Chapter 3.
2.2 The return on the Government Pension Fund
2.2.1 Introduction
A thorough discussion of the return on the Government Pension Fund is provided below. A more detailed analysis of return and risk in the Fund is appended to the present Chapter. The analysis distinguishes between the return resulting from general developments in the markets where the Fund is invested and the return resulting from the investment choices made by Norges Bank and Folketrygdfondet (also known as the National Insurance Scheme Fund). The analysis shows that both the returns on the Fund and the variations in such returns are primarily determined by the general investment strategy of the Pension Fund as laid down by the Ministry. Moreover, the analysis shows that Norges Bank has generated excess return through the management of the Government Pension Fund – Global every year since 1998. Folketrygdfondet’s management performance in respect of the Government Pension Fund – Norway has been more or less in line with the Fund’s benchmark portfolio over the period 1998–2006. Key performance evaluation terms are explained in Box 2.1.
Textbox 2.1 Important performance evaluation terms
For purposes of evaluating the operational management of the Government Pension Fund, its actual return is compared to what could alternatively have been achieved through a fairly similar investment of its funds. It is common to express an alternative investment in the form of a benchmark portfolio . The composition of the benchmark portfolio may be said to reflect the owner’s general choice of investment strategy.
Index management aims for the return on the actual portfolio to closely match the return on the benchmark portfolio. If the indices making up the benchmark portfolio encompass most of the market-traded securities, one will through index management achieve about the same return as the overall market. Return achieved through broad exposure to the securities markets is often termed beta-return in financial literature.
So-called active management involves the manager choosing a composition of the actual portfolio that deviates from the benchmark portfolio. The intent behind such deviations is to achieve an excess return relative to the benchmark portfolio. This excess return is termed alpha-return in financial literature.
The standard deviation is a measure of the fluctuations in returns over a period. The higher is the standard deviation, the more volatility (fluctuations) relative to the average return.
Market risk is the risk that the value of a securities portfolio will change as the result of changes in equity prices, exchange rates and interest rates, cf. Box 3.2.
The normal distribution is a probability distribution frequently used in statistical methods. If it is assumed that the rates of return follow a normal distribution, their probability distribution will be exhaustively characterised by the expected return and the standard deviation of the return. If a security has an expected return of 5 pct. and a standard deviation of 10 pct., the return thereon is expected to fall within the range from -5 to 15 pct. in two out of every three years.
The benchmark portfolio is also used as a risk management tool for operational management purposes, by way of the definition of a limit as to what deviation is accepted between the actual investments and the benchmark portfolio. The difference between actual investments and the benchmark portfolio may be defined as a difference portfolio. The standard deviation of the return on the difference portfolio, which is labelled relative volatility or tracking error , is a measure of the market risk involved in operational management, cf. Box 3.2. The Ministry has fixed a limit of 1.5 pct. for the expected tracking error of the Government Pension Fund – Global. This means, somewhat simplified, that over time the difference between the return on the Fund and the return on the benchmark portfolio will in two out of every three years not exceed 1.5 pct. As far as the management of the Government Pension Fund – Norway is concerned, the Executive Board of Folketrygdfondet has for 2007 fixed an upper limit of 3.5 pct. on expected tracking error. Reference is made to the National Budget for 2002 for a more detailed discussion of tracking error.
The return on the benchmark portfolio is calculated as a weighted sum of the return on the benchmark portfolio for equities and the benchmark portfolio for fixed-income securities, respectively. However, the risk associated with the benchmark portfolio is normally lower than a weighted sum of the risk associated with the equity benchmark and the risk associated with the fixed-income benchmark. The reason for this is that returns in the stock and fixed-income markets are not fully correlated. A measure of the degree of (linear) covariation between rates of return is the correlation coefficient . This is a statistical measure that can range between -1 and +1, with -1 denoting perfect negative correlation and +1 denoting perfect positive correlation. Only in the case of perfect positive correlation will the risk associated with the overall benchmark portfolio of the Fund be equal to the weighted sun of the risks associated with the equity and fixed-income benchmarks. In all other cases, the risk will be lower. The risk reduction achieved by spreading the investments across different assets in this manner is labelled a diversification benefit , cf. Box 2.7.
Table 2.1 shows the real returns, net of management costs, on the two parts of the Government Pension Fund over the years 1997–2006.
Table 2.1 Annual real returns on the Govern,ment Pension Fund – Global1 and the Government Pension Fund – Norway,2 less management costs. 1997–2006. Percent
| Net real return | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 1997–2006 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Pension Fund – Global | 7.15 | 8.20 | 10.93 | 0.35 | -3.66 | -6.62 | 10.75 | 6.30 | 8.46 | 5.55 | 4.58 |
| Pension Fund – Norway | 5.60 | -2.29 | 6.24 | 3.08 | 0.97 | 0.49 | 13.31 | 10.12 | 7.28 | 8.55 | 5.24 |
1 Geometric real return in international currency calculated on the basis of a weighted average of retail price growth in the countries included in the Fund’s benchmark portfolio. Average management costs were 0.09 pct. of the assets under management over this period.
2 Geometric real return in Norwegian kroner. Management costs are assumed, for technical calculation purposes, to have been 0.05 pct. of assets under management, which is higher than historical management costs because a significant portion of the assets managed by Folketrygdfondet was held in the form of sight deposits with the treasury.
Source Norges Bank, Folketrygdfondet and the Ministry of Finance
It follows from Table 2.1 that the annual real return on the Government Pension Fund – Global was 4.6 pct. over the period 1997–2006, as measured in foreign currency. The real return is what is of relevance under the fiscal rule adopted for budget policy purposes, cf. Box 1.1. There are also analyses of historical real returns on equities and bonds, and expectations as to future real returns (as measured in foreign currency), which are used in the investment strategy discussion in Chapter 3. The subsequent analysis of return and risk on the part of the Government Pension Fund in Chapter 2 is based on nominal return data. For purposes of these analyses, which compare, inter alia , management performance to the return on a benchmark portfolio, not much information would have been added by adjusting the return data for inflation. 3
2.2.2 The return on the Government Pension Fund – Global
The Government Pension Fund – Global amounted to NOK 1,783.7 billion as per yearend 2006. The value of the equity portfolio was NOK 725.9 billion, whilst NOK 1,057.8 billion was invested in fixed-income securities. This corresponded to an allocation of 40.7 and 59.3 pct., respectively, of the overall portfolio of the Government Pension Fund – Global. The nominal return on the sub-portfolios of the Government Pension Fund – Global since the beginning of 1998 is illustrated in Figure 2.4.
Figure 2.4 Accumulated nominal return on the sub-portfolios of the Government Pension Fund – Global, as measured in the Fund’s currency basket. Index as per yearend 1997 = 100
Source Norges Bank and the Ministry of Finance
In 2006, the return on the Government Pension Fund – Global was 7.9 pct., as measured in foreign currency when the returns in individual currencies are weighted in accordance with Fund’s benchmark portfolio. The return on the equity portfolio was 17.0 pct., and the return on the fixed-income portfolio was 1.9 pct. When measured in Norwegian kroner, the Fund last year generated a return of 5.9 pct. The difference in return and risk as measured in international currency and in Norwegian kroner is described in more detail in Box 2.6 of the performance analysis. From 1997 until yearend 2006, the average annual return was 6.5 pct. as measured in foreign currency, whilst the real return net of management costs was 4.6 pct., cf. Table 2.1.
The return on the Government Pension Fund – Global depends in large part on market developments in terms of the benchmark portfolio established by the Ministry, cf. Chapter 3. The benchmark portfolio is a broadly based portfolio of equities and bonds that are spread across many regions and sectors. Consequently, the return on the benchmark portfolio reflects general developments in the international securities markets. The performance analysis appended to the present Chapter shows that the decisions relating to the establishment of the Fund’s strategy, as operationalised through the design of the benchmark portfolio, have determined between 90 and 95 pct. of the return on the Fund.
Norges Bank may, in its management of the Fund, deviate from the benchmark portfolio within a risk limit, cf. Box 2.1. The contribution from such active management is measured on an ongoing basis, by way of developments in the value of the Fund being compared to developments in the benchmark portfolio. In 2006, the return on the Pension Fund – Global was 0.15 percentage points higher than the return on the benchmark portfolio. This represents about NOK 2.5 billion. When taken in isolation, the equity portfolio delivered a negative excess return of 0.05 percentage points, whilst the fixed-income portfolio generated a positive excess return of 0.24 percentage points. The actual deviation from the benchmark portfolio in 2006 corresponded to 0.37 pct. tracking error (or relative volatility).
Textbox 2.2 Annual report from Mercer on Norges Bank’s management of the Government Pension Fund – Global in 2006
Since 2003, the Ministry of Finance has had an agreement with the consultancy firm Mercer Investment Consulting for verification and analysis of the return on the Government Pension Fund – Global on a regular basis. Mercer uses Mellon Analytical Solutions (MAS) as a subcontractor as far as the return and value measurements are concerned.1 There are prepared simple monthly reports and a more extensive annual rapport that is published on the website of the Ministry of Finance (www.finansdepartementet.no).
The annual report from Mercer shows that the market value of the Fund and the return data for 2006 are in conformity with Norges Bank’s annual report for 2006. Such very limited deviations as have been identified through the year, are explained in the report. One aims for Mercer’s verification to be as independent from Norges Bank’s computations as possible. MAS therefore receives data directly from Norges Bank’s custodian banks and index providers.2 \ MAS calculates the asset, regional and sectoral weights itself, and performs control calculations of the weights of the rebalancing portfolio. The control procedures are described in more details in the appendix to the report.
Mercer’s annual report contains analyses of the Fund’s return and excess return. Mercer has, inter alia, prepared a style analysis of the equity portfolio, which shows that the portfolio has featured several overweights toward certain management styles. The analysis, which is carried out at the end of each quarter, shows that the equity portfolio has in 2006 been characterised by overweight in “growth companies” and underweight in “value companies”.3 This was also the case in 2005, but the overweight is even more pronounced in 2006. The style analyses also show that the equity portfolio has featured, on an ongoing basis, an overweight in companies with a beta in excess of the average for the benchmark portfolio. Beta is a measure of companies’ covariation with market developments, cf. Box 2.8.
Mercer’s sectoral analysis of the equity portfolio shows that Norges Bank does not establish significant sectoral positions relative to the benchmark portfolio. Furthermore, the analyses show that the Fund has over several years been invested in small, listed limited companies (“small-cap”), which have formed part of the Fund’s investment universe, but not of the benchmark portfolio, cf. the discussion in Chapter 3.3. As per yearend 2006, the equity portfolio was invested in about 3,400 companies, whilst the benchmark portfolio comprised in excess of 2,400 companies. This difference is primarily caused by the Fund’s “small-cap” investments.
Mercer’s report indicates that part of the excess return achieved by Norges Bank can be explained by certain factors, like market developments, “small-cap” and “growth and value companies”.
1 Mellon Analytical Services was previously named CAPS. This company has carried out value and return computations for the Government Pension Fund – Global since 1998.
2 Custodian institutions are usually banks. The most important duty of a custodian institution is to ensure that the securities of each customer are duly registered and “kept” in securities depositories in the various countries wherein which such securities are issued. A custodian performs important functions in the settlement of securities trades, and also provides accounting services. Moreover, various supplementary services may be agreed, like for example the monitoring of external managers and the lending of securities.
3 Listed equities are often grouped into two categories; value and growth companies, based on several key financial ratios like turnover growth, market value relative to earnings or book value, etc.
When evaluating the performance of an asset manager, it is customary to adopt a time horizon of several years, and the Ministry has in its performance analysis examined both the excess return over the last three years and over the period 1998–2006, i.e. the entire period during which the Fund has been permitted to invest in equities. On average over the period 2004–2006, the excess return on the Government Pension Fund – Global was 0.58 percentage points, whilst the average excess return on the Fund since 1998 was 0.48 percentage points. Since the beginning of 1998, equity management has contributed about two thirds of the overall excess return. Total accumulated excess return was NOK 29 billion. The annual tracking error of the Government Pension Fund – Global over the same period was 0.38 pct.
Textbox 2.3 Comparison of the returns and costs associated with the management of the Government Pension Fund – Global and those associated with other pension funds
The Ministry of Finance has commissioned CEM Benchmarking Inc. (“CEM”) to prepare an extensive annual rapport that compares the returns, excess returns and management costs associated with the Pension Fund – Global and those associated with other large funds. The reference group comprises the largest funds in the CEM survey (i.e. 14 US, 2 Canadian and 3 European funds). Since the Pension Fund holds a portion of investments in Europe that is well above the average for this group, the return on the Fund is also compared to the return on 23 European funds. The median size of European funds is about NOK 150 billion, as compared to about NOK 680 billion for the reference group. The data set is based on self-reporting from the funds that purchase services from CEM.
Excess return
A comparison over time of the actual return on the Government Pension Fund – Global and the return on the benchmark portfolio shows the excess return generated by Norges Bank. Figure 2.5. shows the average excess return on, and tracking error of, the Government Pension Fund – Global and the various reference groups over the 5-year period 2001– 2005. It follows from the Figure that the Pension Fund – Global has, like the majority of the other funds, achieved positive excess returns over this period. It may also be noted that the Pension Fund – Global differs from most other funds inasmuch as there is less variation in its excess return, i.e. less tracking error.
Figure 2.5 Average annual excess return on, and tracking error of, the Government Pension Fund – Global and other funds. 2001– 2005. Percent.
Source CEM Benchmarking Inc.
CEM has calculated that the Pension Fund – Global achieved, over the 5-year period until 2005, an average annual excess return of 0.5 percentage point. In comparison, the most typical excess return (the median) was 0.5 percentage point for the European funds and 0.6 percentage point for the reference group. At the same time, the analyses show that the excess return on the Pension Fund – Global has been generated by assuming considerably less active risk (tracking error) than the other funds.
Both the reference group and the European funds have generated excess returns in asset classes in which the Pension Fund – Global is not invested. For the chosen 5-year period, the excess return has been especially high within real estate. This means that Norges Bank’s excess return is better, relative to that of comparable funds, than can be inferred directly from the CEM report.
Management costs
The CEM report shows that Norges Bank’s overall management costs in 2005 for the Government Pension Fund – Global were 0.11 pct. of the average capital under management. These are the lowest management costs of all the funds from which CEM gathers data. Since the overall management costs in large part reflect the asset structure of the funds, which as far as the Pension Fund – Global is concerned follows from the Ministry of Finance’s investment strategy, such comparisons do not fully clarify whether Norges Bank’s management effort is cost effective.
CEM has therefore prepared a cost benchmark based on the asset structure of the Pension Fund – Global. The cost benchmark indicates which costs the reference group – the world’s largest pension funds – would have incurred with the same asset structure as the Pension Fund. The analysis shows that actual management costs in 2005 were 0.03 percentage point less than the cost benchmark. This is primarily caused by Norges Bank having chosen more internal management than has the reference group. Internal management is considerably less expensive than external management.
The return on the benchmark portfolio
The Ministry of Finance has established a benchmark portfolio for the Government Pension Fund – Global. The analyses of CEM show that the average annual return on the benchmark portfolio of the Pension Fund – Global over the 5-year period until 2005 was 4.3 pct. as measured in the currency basket of the benchmark portfolio. Correspondingly, the median return was 4.5 pct. and 5 pct. for the reference group and the European funds, as measured in USD and Euros, respectively.
Comparison of aggregate returns between funds is difficult because different funds hold assets denominated in different currencies and have different benchmark currencies. Exchange rate fluctuations imply that the return will depend on the benchmark currency one uses. For example, the return on the Fund’s benchmark portfolio was 2.0 pct. as measured in Euros, whilst it was 6.8 pct. as measured in USD, over the 5-year period until 2005. The large difference was caused by a significant depreciation of USD against Euros over this period.
Differences in the returns on the benchmark portfolios will also result from differences in terms of asset classes and regional composition. The Pension Fund – Global has thus far differed from other large pension funds by way of a lower equity portion and by way of the Fund not being invested in real estate and unlisted equities. CEM has calculated that if the other funds had featured the same assets class composition as the Pension Fund over the 5-year period until 2005, their annual indexed return would have been reduced by 0.8 percentage point for the reference group and 0.5 percentage point for the European funds. This is primarily attributable to the fact that the Pension Fund has not been invested in real estate, whilst the other funds have held an average real estate allocation of 8 pct. This 5-year period has been a historically good period as far as the property market is concerned. The Pension Fund – Global also differs from other funds inasmuch as its investments are spread across many markets and currencies, whilst most of the pension funds in the reference group hold the main part of their investments in their domestic markets.
The investment strategy underpinning the composition of the benchmark portfolio is based on trade-offs between long-term expectations concerning return and risk in the capital markets, cf. Chapter 3. In such a perspective, a 5-year period is short. If the comparisons had been made for a different 5-year period, the findings might have been different.
A summary of the analyses for 2005 is published on the Internet (www.finansdepartementet.no)
The risk associated with the Government Pension Fund – Global, as measured by the annual standard deviation of returns, has since 1998 been 5.3 pct. as measured nominally in local currency, cf. the appended performance analysis. Variations in the return on the benchmark portfolio have explained more than 99 pct. of the variations in the return on the Fund. Consequently, Norges Bank’s active management has made almost no contribution towards increasing risk on the part of the Fund, cf. Box 2.1. The relationship between the average excess return generated by Norges Bank and the increase in risk resulting from such investment choices, indicates good performance as far as Norges Bank’s active management is concerned.
Textbox 2.4 Comparison of the excess return on the Government Pension Fund – Global and the excess returns on other funds
WM Performance Services (“WM”) has for 2006 prepared a report that compares the excess return on the Government Pension Fund – Global to the excess returns on pension funds and life insurance companies domiciled in the Netherlands and the United Kingdom. As per yearend 2006, the aggregate market value of the funds was NOK 12,075 billion, and their average market value was NOK 31.5 billion. In comparison, the market value of the Pension Fund – Global was NOK 1,572 billion in 2006, calculated as a monthly average.
Figure 2.6. shows the average excess returns on, and tracking error of, the Government Pension Fund – Global and other funds over the last three years. The Figure also shows the findings for the 15 and 50 largest funds. In 2006, these two groups had an average market value of NOK 337 and 33 billion, respectively.
The Figure shows that the Pension Fund – Global has had a positive excess return, and distinguishes itself from the majority of other funds by less variation in its excess return, i.e. a lower tracking error. WM points out that the excess returns on the large funds are more concentrated around zero, and their explanation for this is that when active management is spread across more assets and more types of management, the probability of large deviations from the return on the benchmark portfolio is reduced.
Figure 2.6 Average annual excess return on, and tracking error of, the Government Pension Fund – Global and other funds. 2004–2006. Percent
Source WM Performance Services.
WM has calculated that the most typical excess return (the median) for all funds included in the survey is zero for both the 3-year and the 5-year period until yearend 2006. The excess return on the Government Pension Fund – Global has been close to those on the 25 pct. best funds for both of these periods, whilst the tracking error has been amongst the 5 pct. lowest observations. This results, according to WM, in a very attractive and unusual trade-off between excess return and tracking error.
The ratio between excess return and tracking error is often referred to as the information ratio (IR). WM points out that care should be exercised in interpreting IR. If tracking error is very low, even minor return differences relative to the return on the benchmark portfolio may result in a high IR, either negative or positive. WM writes that an IR of 0.5 is often characterised as good, and points out that the Pension Fund – Global has over several years achieved an IR in excess of 1, which is unusual.
In the comparison with other funds, WM points out that most large funds have a capital inflow of less than 5 pct. of their opening value during the course of a year, whilst the Pension Fund – Global in 2006 received more than 20 pct. of its value as per the beginning of the year. WM Performance Services writes that having to invest that much capital, both in absolute terms and relative to the size of the Fund, makes it more challenging to generate excess return.
The report from WM also contains a comparison of the excess returns on the equity and fixed-income portfolios of other funds. In the main, these analyses yield the same conclusions as those pertaining to the Fund as a whole. The report for 2006 is published on the Internet (www.finansdepartementet.no).
The Ministry receives reports from Mercer Investment Consulting, CEM Benchmarking Inc. and WM Performance Services regarding the management of the Government Pension Fund – Global on annual basis, cf. Boxes 2.2 – 2.4. The Ministry has commissioned Mercer to verify and analyse the return on the Pension Fund – Global. In the annual report for 2006, Mercer shows that its independent computations of return and market value for the Fund are in conformity with the computations of Norges Bank. The return analyses indicate, inter alia , that parts of the excess return achieved by Norges Bank may be explained by certain factors, like market development, returns on small-cap stocks and on “growth and value” stocks. The Ministry of Finance will on the basis of, inter alia , these findings, refine factor analyses in the effort to evaluate the management of the Government Pension Fund.
Comparison of the performance of the Pension Fund – Global and that of other large institutional managers forms an important part of the analyses of Norges Bank’s active management. According to the analyses from CEM, the excess return on the Pension Fund – Global is in line with that of comparable funds, whilst the analyses from WM Performance Services conclude that the excess return is higher than that of other funds. Both analyses show that the tracking error is clearly less than that of other funds. Such comparisons are difficult because the scope for generating excess return will vary with the investment limits and guidelines laid down in respect of the funds. Both the Pension Fund – Global and other funds have benchmark portfolios that do not include all the types of assets in which the funds may invest. Normally, it is easier to achieve excess return relative to such a benchmark than if one were to measure returns against a more comprehensive benchmark that more accurately reflects developments in the value of all types of assets in which the Fund may invest.
It is also difficult to find the most relevant reference group. The data sets used by both WM Performance Services and CEM are extensive, but different. For example, CEM does not cover funds in the United Kingdom, whilst these account for the bulk of the funds encompassed by the data set of WM Performance Services. A second difference is that most of the world’s large pension funds are included in the CEM study. Consequently, in terms of market value the CEM funds are larger and more comparable to the Pension Fund – Global. A third difference is that more than 25 pct. of the funds covered by WM Performance Services are life insurance companies, whilst this type of fund is not encompassed by the CEM study. Different types of funds may perform differently. For example, a CEM survey shows that US pension funds generated higher excess returns than did US mutual funds between 1992 and 2004. 4
A peculiar characteristic of the Pension Fund – Global, as compared to other funds is its large inflow of new capital. WM Performance Services points out that most large funds have an inflow of less than 5 pct. of its opening value during the course of one year, whilst in 2006 the Pension Fund – Global registered an inflow in excess of 20 pct. of its value as per the beginning of the year. WM Performance Services writes that being charged with investing this much capital, in both absolute terms and relative to the size of the Fund, makes it more demanding to generate excess return.
2.2.3 The return on the Government Pension Fund – Norway
The value of the Government Pension Fund – Norway was NOK 106.9 billion as per yearend 2006. The value of the equity and fixed-income portfolio was NOK 63.4 and 43.5 billion, respectively, as per yearend. 5 This corresponded to an allocation of 59.3 and 40.7 pct., respectively, of the overall portfolio of the Government Pension Fund – Norway. The nominal return on the sub-portfolios of the Government Pension Fund – Norway since the beginning of 1998 is illustrated in Figure 2.7.
Figure 2.7 Accumulated nominal return on the Government Pension Fund – Norway’s sub-portfolios, as measured in Norwegian kroner. Index as per the beginning of 1997 = 100
Source Folketrygdfondet and the Ministry of Finance
The return on the Government Pension Fund – Norway was 11.7 pct. in 2006. The return on the Fund’s Norwegian and Nordic equity investments was 32.1 pct. and 24.1 pct., respectively, whilst the return on the fixed-income investments was 1.3 pct. Between 1997 and yearend 2006, the average annual return was 7.1 pct.
The Ministry has defined general investment limits for the Government Pension Fund – Norway, whilst the more detailed benchmark portfolio of the measurement of return and risk is established by the Executive Board of Folketrygdfondet, cf. the discussion in Chapter 5.3. The return on the Government Pension Fund – Norway, which is primarily invested in the Norwegian securities market, is highly dependent on market developments as far as the benchmark portfolio is concerned.
The discrepancy between the return on the actual portfolio and that on the benchmark portfolio can mainly be attributed to the investment decisions of Folketrygdfondet. In 2006, the return on the Pension Fund – Norway was 1.13 percentage points higher than the return on the benchmark portfolio. The excess return mainly relates to the Fund having exposed the portfolio to less interest rate risk (duration) during a period of increasing market interest rates, and having held a higher portion of equities than suggested by the benchmark index, cf. Box 3.2. When taken in isolation, the fixed-income portfolio achieved an excess return of 0.62 percentage points last year, whilst the Norwegian and the Nordic equity portfolio generated a negative excess return of 0.36 and 1.14 percentage points, respectively. The actual deviation from the benchmark portfolio in 2006 corresponded to 0.8 pct. tracking error. The Pension Fund’s fixed-income portfolio has changed considerably in character since 2006, as the result of the termination of the sight deposits with the tresury, and the Ministry is aware that Folketrygdfondet has made changes to its operational management structure for Nordic equities with a view to improving performance.
The Government Pension Fund – Norway adopts a long investment horizon. The Fund is a long-term investor, and will hold a large portfolio of equities in the Norwegian market, the composition of which Folketrygdfondet must examine. In line herewith, the Ministry focuses on performance developments over time in its follow-up of Folketrygdfondet’s performance.
Over the period 1998–2006, the performance of Folketrygdfondet has on the whole been satisfactory, and more or less in line with the benchmark portfolio of the Pension Fund – Norway. Positive excess return has been generated through the management of Norwegian equities, whilst negative excess returns have resulted from fixed-income management and the management of Nordic equities. 6 Over the three-year period 2004–2006, the annual negative excess return on the overall portfolio has been 0.97 percentage points on average. Annual tracking error over the periods 1998–2006 and 2004–2006 was 1.3 pct., cf. the appended performance analysis.
When taken in isolation, the Norwegian equity portfolio has generated a positive average annual excess return of 1.0 percentage point over the period 1998–2006, whilst this portfolio has registered negative average annual excess return over the period 2004–2006 to the order of 3.62 percentage points. The performance analyses set out in the Appendix to the present Chapter shows that the Norwegian equity portfolio of Folketrygdfondet tends to outperform the benchmark portfolio during periods of weak performance on the Oslo Stock Exchange, and to underperform somewhat relative to the benchmark portfolio during those periods characterised by a strong upturn in equity prices. The analysis also shows that the variations in the return on the benchmark portfolio explain about 93 pct. of the variations in the return on the Fund. The risk associated with the Government Pension Fund – Norway, as measured by the annual standard deviation of returns, has been 4.3 pct. since 1998, as measured nominally in Norwegian kroner.
In the same manner as in its follow-up of Norges Bank’s management, the Ministry will be contemplating the use of external consultants to review various aspects of Folketrygdfondet’s management of the Government Pension Fund – Norway.
Following the termination of the sight deposits with the treasury, the equity portfolio accounts for a higher portion of the overall portfolio than was previously the case. Consequently, one needs to be prepared for more pronounced annual fluctuations in the overall return on the Pension Fund – Norway in coming years than have been experienced thus far.
2.3 Management costs
The management agreement between the Ministry of Finance and Norges Bank in respect of the Pension Fund – Global lays down principles governing the compensation to be received by Norges Bank in return for managing the Fund. According to the agreement, the compensation payable to Norges Bank shall cover the actual management costs, up to a maximum limit, which for 2006 has been fixed at 0.1 pct. of the average market value of the Fund. The maximum compensation limit is determined on the basis of, inter alia , information on the costs associated with this type of management in pension funds of corresponding size. The Ministry of Finance commissions the Canadian company CEM Benchmarking Inc. to prepare the analyses on which the cost comparisons are based. In addition to the reimbursement of costs up to the maximum limit, Norges Bank is compensated for such part of the fees of external managers as are incurred as a result of the excess return achieved.
Management costs for 2006, exclusive of performance-related fees, were NOK 1,138 million. This represented an increase of 24 pct. over 2005. The average size of the Fund increased by 33 pct., thus implying that costs measured as a share of the average portfolio were reduced from 0.08 pct. in 2005 to 0.07 pct. in 2006. Consequently, management costs exclusive of performance-related fees are well below the maximum limit. Inclusive of performance-related fees to external managers, the costs amounted to NOK 1,526 million, which is 0.1 pct. of the average market value when calculated as an annual rate.
The costs incurred by Folketrygdfondet in its management of the Government Pension Fund – Norway represent (following the termination of the sight deposits with the treasury) about 0.05 pct. of the assets under management. The management costs associated with the Government Pension Fund – Norway are not entirely comparable to the costs associated with the management of the Government Pension Fund – Global. Norges Bank has opted to use external management mandates, which when taken in isolation are more expensive than internal management. Furthermore, the asset management carried out by Norges Bank is more extensive because, inter alia , the assets of the Pension Fund – Global are spread across many more countries and companies than those of the Pension Fund – Norway. On the other hand, asset management is subject to economies of scale, and Norges Bank is amongst those benefiting therefrom.
The Ministry intends to enter into a management agreement with Folketrygdfondet that would lay down the principles governing the compensation to be received by Folketrygdfondet in return for managing the Government Pension Fund – Norway, cf. the discussion in Chapter 5.
2.4 More detailed analysis of return and risk in the Government Pension Fund (Appendix)
II.1 Introduction
In the analysis of the return on the Government Pension Fund, a distinction is made between the return resulting from general developments in the markets where the Fund is invested, and the return resulting from the investment choices made by Norges Bank and Folketrygdfondet. Norges Bank’s management of the Government Pension Fund – Global is premised on the general investment strategy as expressed in a benchmark portfolio established by the Ministry. The strategy of the Government Pension Fund – Norway is defined by the Ministry in the form of general limits on investments in equities and fixed-income securities, whilst the detailed benchmark portfolio of the measurement of return and risk is established by the Executive Board of Folketrygdfondet. The analysis below demonstrates that the Ministry’s decisions as to the investment strategy of the Funds are the main determinants of both the returns on the Funds and the variations in such returns.
In their management of the Government Pension Fund, Norges Bank and Folketrygdfondet may deviate from the benchmark portfolio for purposes of generating excess return, within defined risk limits. The analysis shows that Norges Bank has generated excess return through its management of the Government Pension Fund – Global every year since 1998. Folketrygdfondet’s management performance in respect of the Government Pension Fund – Norway has been more or less in line with the Fund’s benchmark portfolio over the period 1998–2006.
Out of concern for the statistical analyses in the performance discussion below, the average return is in the present Appendix to Chapter 2 calculated as an arithmetic average based on monthly return data. The conversion to annual returns is done by multiplying by a factor of 12. In the main part of Chapter 2, the return was calculated geometrically, in line with the return data reported in the annual reports of Norges Bank and Folketrygdfondet concerning the management of the Government Pension Fund – Global and the Government Pension Fund – Norway, respectively. Chapter 3 also uses a geometrically calculated average return, since this is often deemed to better reflect the long-term expected growth rate of an investment, cf. Box 2.5.
Textbox 2.5 On average real return and its relationship to the real value of the Fund
There are different measures for an average value. The two most common measures are the arithmetic and the geometric average. A simple numerical example illustrates the difference between them. Assume that an investment of NOK 100 has after one year doubled its value to NOK 200 (up 100 pct.), and has thereafter been halved the subsequent year, to NOK 100 (down 50 pct.). The arithmetic average return will in such a case be 25 pct., whilst the geometrically calculated average return will be 0 pct.1 Consequently, the arithmetic average represents the mean value of the returns, whilst the geometric average denotes the average growth rate of the investment.
It can be demonstrated that the geometric average is always less than, or as a special case identical to, the arithmetic average. Which average is the most relevant one will depend on the purpose of the calculations. In projections of long-term expected annual percentage real returns it is common to focus on a geometric average return, which also best expresses the historical return. In its quarterly and annual report, Norges Bank reports the real return on the fund capital as a geometric average.
The economic policy guidelines imply that the petroleum revenues are phased gradually into the economy, more or less in line with the expected real return on the Global part of the Government Pension Fund. The fiscal rule assumes that a real return of 4 pct. represents a reasonable expectation as to the annual percentage return over time for the Pension Fund – Global, cf. Report No. 29 (2000-2001) to the Storting. One intention behind the policy rule is to preserve the real value of the fund capital. If the expected geometric average real return on the Fund is close to four pct., one will more or less preserve the real value of the Pension Fund through annual withdrawals equal to 4 pct. of the Fund’s value.
1 The arithmetic average is calculated as (100 % - 50 %)/2 = 25 %. The geometric average is calculated as ((1+1)(1-0.5))1/2 -1=0.
Sub-chapters II.2 – II.4 analyse the return on the Government Pension Fund – Global in more detail, whilst the analysis of the return on the Government Pension Fund – Norway is to be found in Sub-chapters II.5 – II.7.
II.2 Developments in the benchmark portfolio of the Government Pension Fund – Global
The return on the Government Pension Fund – Global depends in large part on market developments in the benchmark portfolio established by the Ministry. The benchmark portfolio is a broadly composed portfolio of equities and bonds that is spread across many regions and sectors. Consequently, the return on the benchmark portfolio reflects general developments in the international securities markets. Indices from the providers FTSE and Lehman Brothers are adopted for equities and bonds, respectively.
Table 2.2 shows the return on the Fund’s benchmark portfolio in local currency. The difference between returns in local currency and returns in Norwegian kroner is explained in Box 2.6. As illustrated in Table 2.2, returns have been both considerably higher and less variable in recent years than over the period 1998 – 2006 as a whole.
Table 2.2 Average annual return and annual standard deviation of the return on the benchmark portfolio of the Government Pension Fund – Global, measured nominally in local currency. Percent
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| The benchmark portfolio | |||
| Average return per year | 5.72 | 8.44 | 7.54 |
| Standard deviation | 5.18 | 3.18 | 3.13 |
| The equity benchmark | |||
| Average return per year | 7.24 | 15.60 | 16.14 |
| Standard deviation | 14.39 | 7.17 | 7.40 |
| The fixed-income benchmark | |||
| Average return per year | 5.04 | 3.58 | 1.70 |
| Standard deviation . | 3.13 | 2.67 | 2.63 |
Source Norges Bank and the Ministry of Finance.
Textbox 2.6 Fund return measured in Norwegian kroner and in local currency
A country’s savings are the sum of fixed capital formation domestically and the surplus of the current account of the balance of payments, i.e. net financial investments abroad. The surplus on the current account corresponds to the sum of net financial investments in the various domestic sectors. Consequently, individual sectors in Norway may increase their financial assets and liabilities as against other domestic sectors. However, for the country as a whole, financial assets can only be accumulated as against other countries. In Norway, the State accounts for a considerable share of these savings, by way of its investment of a large share of the petroleum revenues abroad through the Government Pension Fund – Global.
The fund capital and the returns thereon enable the funding of increased net imports, thereby freeing up factors of production from businesses exposed to international competition. How large imports can be funded on the basis of the Fund depend only on the foreign exchange holdings of the Fund, and not on the value of the Fund in Norwegian kroner. The allocation of the Fund’s investments between Europe, the Americas and Asia is based on, inter alia , Norway’s import weights, for reasons of preserving the international purchasing power of the Fund.
The return on the Government Pension Fund – Global may be measured in both Norwegian kroner and in the Fund’s currency basket.1 Since the international purchasing power of the Fund remains unaffected by changes to the Norwegian krone exchange rate, the focus in the performance reporting of the Fund is often on the return as measured in the Fund’s currency basket (or, in other words, in local currency). Figure 2.8 shows the monthly return on the benchmark portfolio of the Government Pension Fund – Global as measured in Norwegian kroner and in local currency.
Figure 2.8 Return on the benchmark portfolio for a given month as measured in Norwegian kroner and in local currency. 1998–2006. Percent
Source Norges Bank and the Ministry of Finance
In the same way that the return on the Fund may be measured in both Norwegian kroner and in local currency, the variations in such returns may be measured in Norwegian kroner and in local currency. The variations in returns as measured in local currency are the important measure for purposes of assessing the risk associated with developments in the Fund’s purchasing power. The variations in returns as measured in NOK have been higher than the variations in returns as measured in local currency, cf. Figure 2.9. This is because variations in returns in NOK are caused by both variations in returns in local currency and by fluctuations in the Norwegian krone exchange rate. If the fluctuations in the Norwegian krone exchange rate had been perfectly correlated with returns in the international markets, the standard deviation of the return in NOK would have been equal to the sum of the standard deviation of returns in local currency and the standard deviation of the fluctuations of the Norwegian krone exchange rate. Computations show that the correlation (the covariation) between returns in Norwegian kroner and returns in the local securities markets is close to nil. This implies that the variation in the return as measured in NOK has been considerably less than the sum of the two standard deviations. The standard deviation of the fluctuations in the Norwegian krone exchange rate has been less than 7 pct. over the period 1998 – 2006. It follows from Figure 2.9 that the risk (the volatility) as measured in Norwegian kroner is considerably higher than the risk as measured in local currency, and that the difference increases during those periods when exchange rate volatility is higher.
Figure 2.9 Rolling twelve-month standard deviation of returns on the benchmark portfolio, as measured in Norwegian kroner and in local currency. 1998–2006. Percent
Source Norges Bank and the Ministry of Finance
Table 2.3 shows the average annual return and the standard deviation of returns on the actual portfolio, the benchmark portfolio and the difference portfolio, as measured in Norwegian kroner and in local currency.
Table 2.3 Average annual return and annual standard deviation of returns on the actual portfolio, benchmark portfolio and difference portfolio of the Government Pension Fund – Global, as measured in Norwegian kroner and in local currency. 1998–2006. Percentages and percentage points
| Actual portfolio | Benchmark portfolio | Difference portfolio | |
|---|---|---|---|
| Returns in Norwegian kroner (1) | |||
| Average return per year | 6.13 | 5.67 | 0.46 |
| Standard deviation | 8.51 | 8.45 | 0.38 |
| Returns in local currency (2) | |||
| Average return per year | 6.18 | 5.72 | 0.46 |
| Standard deviation | 5.29 | 5.18 | 0.38 |
| Return difference between (1) and (2) | |||
| Average return per year | -0.05 | -0.05 | |
| Standard deviation | 3.22 | 3.26 |
Source Norges Bank and the Ministry of Finance
It follows from the Table that there has over the period 1998 to 2006 been little difference in average annual returns as measured in Norwegian kroner and in local currency. However, the standard deviation of the actual portfolio and of the benchmark portfolio as measured in Norwegian kroner has been about 3 percentage points higher than the standard deviation in local currency.
The difference in volatility as measured in Norwegian kroner and in local currency is more pronounced for the fixed-income portfolio than for the equity portfolio. This has to do with the volatility in the stock markets as measured in local currency having been markedly higher than the volatility in the fixed-income markets. The volatility of the return as calculated in Norwegian kroner for the equity benchmark portfolio is driven by the volatility in the local stock markets to a much greater extent than what is the case for the benchmark portfolio for fixed-income securities. For the latter portfolio, variations in the Norwegian krone exchange rate has had a much greater impact on variations in returns as measured in NOK. The variation in the Norwegian krone exchange rate explains, based on somewhat approximate computations, about 60 pct. of the overall variations in returns on the benchmark portfolio of the Fund, as measured in Norwegian kroner. The corresponding figures for the equity and fixed-income benchmark portfolios are about 20 and 80 pct., respectively.
It follows from Table 2.3 that the return difference (or excess return) and the relative risk as measured in NOK has not been markedly different from the excess return and the relative risk as measured in local currency. This shows that the active management effort of Norges Bank has not been exposed to changes in the value of the Norwegian krone relative to the Fund’s currency basket. However, this analysis does not show whether the active management effort has resulted in an actual portfolio with a different foreign exchange exposure within the investment universe than that of the benchmark portfolio.
1 The return measured in Norwegian kroner is influenced by how the Norwegian krone exchange rate develops relative to the currencies in which the Fund is invested. If the Norwegian krone exchange rate depreciates, on average, as against these currencies, the return on the Fund as measured in NOK will be higher than the return on the Fund as measured in the Fund’s currency basket. This is because there will arise, when converting the foreign exchange value of the Fund to Norwegian kroner, a gain as measured in NOK as the result of the depreciation of the Norwegian krone.
The benchmark portfolio of the Government Pension Fund – Global comprises two benchmark sub-portfolios; one covering the international stock markets and one covering the international fixed-income markets. Equities account for 40 pct. of the benchmark portfolio of the Fund, whilst fixed-income securities account for 60 pct., cf. Chapter 3.
The very high return on the benchmark portfolio for equities in recent years reflects the international increase in equity prices. Fluctuations in the stock market have been abnormally low during this period. Calculations based on return sets that cover long time periods, indicate that the normal volatility of the stock market is in the range of 14 to 16 pct. (as measured by the standard deviation of the return). However, the benchmark portfolio for fixed-income securities has registered relatively low returns in recent years, as compared to normal returns in the fixed-income markets and as compared to returns in the stock markets. Whilst returns in the fixed-income markets have been low, there has been little return volatility.
The return on the Fund’s benchmark portfolio is calculated as the sum of the return on the equity benchmark, as weighted by the equity portion (40 pct.), and the return on the fixed-income benchmark, as weighted by the fixed-income portion (60 pct.). However, the risk of the benchmark portfolio is less than a weighted sum of the risk associated with the equity benchmark and the risk associated with the fixed-income benchmark. The reason for this is that there has been a relatively low degree of covariation between returns in the stock and fixed-income markets. This effect is classified as a diversification benefit, cf. Box 2.7.
Textbox 2.7 Calculation of the diversification benefit for the Government Pension Fund – Global
The benchmark portfolio of the Government Pension Fund – Global comprises equities and bonds. The overall risk associated with the portfolio is determined by the risk associated with each benchmark sub-portfolio and by the degree of covariation between the return in the equity and bond markets. The first bar in Figure 2.10 shows what risk would have been associated with the Fund if returns in the stock and bond markets had been perfectly correlated. The second bar shows the magnitude of the diversification benefit, whilst the third one shows the realised risk associated with the Fund. The diversification benefit has been in excess of 2 percentage points.
Figure 2.11 emphasises the importance of the correlation between returns in the stock and fixed-income markets in terms of the overall risk associated with the benchmark. The historical standard deviation of the overall benchmark has been about 5.2 pct., whilst the correlation between returns in the stock and fixed-income markets over the period 1998-2006 was about -0.4. Based on this correlation it would follow from the graph that the risk associated with the overall benchmark is about 5.3 pct., which provides a good match with historical experience. Under the assumption of perfect positive correlation, the risk associated with the overall benchmark is calculated to be 7.6 pct. Consequently, the difference of 2.3 percentage points indicates the diversification benefit that has been achieved during the period, cf. Figure 2.10.
Figure 2.10 Annual standard deviation for equities, bonds and the Government Pension Fund – Global as a whole. 1998-2006. Local currency. Percent
Source Norges Bank and the Ministry of Finance
Figure 2.12 shows that there are relatively large variations in the correlation between returns in the equity and fixed-income markets over time. This means that the diversification benefit also varies over time. Historically, the correlation between equity and bond returns has been weakly positive for longer return sets, which reduces the diversification benefit somewhat compared to the last ten years.
Figure 2.11 The overall risk associated with the Fund as a function of the correlation between equity and fixed-income returns. Percent1
1 The standard deviation of returns on the Fund’s benchmark portfolio is a function of the standard deviations of returns on the equity and fixed-income benchmarks, the correlation between returns on the equity and fixed-income benchmarks, and the portion of equities in the overall benchmark. The standard deviations of returns on the equity and fixed-income benchmarks are based on historical data for purposes of drawing the Chart. An equity portion of 40 pct. has been assumed, which corresponds to the current equity portion. The risk associated with the Fund can then be calculated as a function of the correlation between returns on the equity and fixed-income benchmarks. Correlation coefficients will always fall into the range from minus one to plus one.
Source Norges Bank and the Ministry of Finance
II.3 Developments in the actual portfolio of the Government Pension Fund – Global
The actual portfolio deviates somewhat from the benchmark portfolio. The deviations mainly reflect active investment decisions on the part of Norges Bank. The Ministry of Finance has established a limit defining the maximum permitted deviation, cf. Box 2.1. The objective of Norges Bank is to achieve a higher return on the actual portfolio than that on the benchmark portfolio, within the risk limit laid down by the Ministry. The return on, and the risk associated with, the actual portfolio will therefore deviate somewhat from those of the benchmark portfolio. The Fund’s return and risk are presented in Table 2.4.
Table 2.4 Average annual return and annual standard deviation of the return on the Government Pension Fund – Global, measured nominally in local currency. Percent
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| Actual portfolio | |||
| Average return per year | 6.18 | 8.99 | 7.69 |
| Standard deviation | 5.29 | 3.36 | 3.27 |
Source Norges Bank and the Ministry of Finance.
It follows from Tables 2.2 and 2.4 that the average return on the actual portfolio is somewhat higher than that on the benchmark portfolio, whilst the risk figures for the two portfolios are about the same. The excess return on the Fund, which is the difference between the return on the actual portfolio and that on the benchmark portfolio, is discussed in more detail in Sub-chapter II.4.
The high degree of correlation between the return and risk associated with the Fund and the return and risk associated with the benchmark portfolio is illustrated in Charts 2.13 and 2.14. Both the level of returns and the variations in returns are about the same for the two portfolios.
Figure 2.12 Rolling twelve-month correlation between the equity and fixed-income return on the benchmark portfolios. 1998–2006.
Source Norges Bank and the Ministry of Finance
Figure 2.13 Returns on the Government Pension Fund – Global and on the Fund’s benchmark portfolio. Monthly return data 1998–2006. Percent
Source Norges Bank and the Ministry of Finance
Figure 2.14 Rolling twelve-month standard deviation of the return on the Government Pension Fund – Global and on the Fund’s benchmark portfolio. 1998–2006. Percent
Source Norges Bank and the Ministry of Finance
These analyses show that both the return and the variation in the Fund’s return are mainly determined by the decisions relating to the establishment of the general investment strategy of the Fund, as operationalised through the formulation of the benchmark portfolio. Between 90 and 95 pct. of the return on the Fund may be attributed to the choices made by the Ministry in designing the investment strategy of the Fund, whilst the remaining 5-10 pct. may be attributed to the investment choices made by Norges Bank within the guidelines laid down by the Ministry. Moreover, the correlation between the return on the Fund and the return on the benchmark portfolio has been very close to 1 (perfect positive correlation), and the variations in returns on the benchmark portfolio have explained more than 99 pct. of the variations in returns on the Fund.
In Figure 2.15, the information in Figure 2.13 is organised differently. Here, the deviations are not sorted by when the return arose. Each dot shows the return on the benchmark portfolio and on the actual portfolio for the same month, as sorted by the level of return. The dot at the north-eastern end of the Figure signifies the return on the Fund and on the benchmark portfolio in May 2003, whilst the dot at the south-western end signifies the same in relation to August 1998. The intervening dots represent combinations of returns on the benchmark portfolio and on the actual portfolio for the other months over the period 1998–2006.
Figure 2.15 Correlation between the return on the Government Pension Fund – Global and on the Fund’s benchmark portfolio. 1998–2006. Percent
Source Norges Bank and the Ministry of Finance
These combinations of rates of return are clustered quite closely along a straight line, as drawn in the Chart. Somewhat simplified, one may say that the straight line has been located such as to minimise the “distance” between the dots and the line. The slope of the line is termed the “beta” of the portfolio whilst the intercept with the y-axis is termed the “alpha” of the portfolio. 7 The beta of the Fund is calculated as 1.018. In statistical terms, this estimate is significantly higher than 1. This implies that there would be a low probability of observing this combination of returns on the actual portfolio and returns on the benchmark portfolio if the beta of the Fund had been 1, cf. Box 2.8.
Textbox 2.8 Expected return and beta exposure
It is commonly assumed that market players are risk averse. This means that they will for a given return choose the investment alternative carrying the lowest risk. In efficient markets, the prices of securities will reflect the risk aversion of investors. The expected return in the stock market is, for example, higher than that in the bond market. The difference in expected returns reflects the difference in risk and the risk aversion of investors.
The beta of a portfolio is a measure of the systematic risk associated with such a portfolio. Systematic risk is the part of the risk that relates to developments in a broadly composed and well-diversified securities portfolio. The unsystematic risk is the part of the risk that is diversified away by investing in a large number of securities. Because investors may relieve themselves of the unsystematic risk at virtually no additional cost, they receive no compensation in the form of higher expected returns when carrying such risk. However, the systematic risk does not disappear. When investing in the securities markets, one will always be exposed to changes in value as the result of general market fluctuations. Higher systematic risk is compensated by way of higher expected returns.1
The risk measure beta is constructed in such a manner that a beta value of one represents the average systematic risk in the market. The benchmark portfolio of the Government Pension Fund is composed of representative market indices. It will therefore, by definition, have a beta of 1. A well-diversified portfolio with a beta in excess of one will on average experience a greater variation in returns than will a portfolio with a beta of 1. The expected return thereon will be correspondingly higher. If the beta of a portfolio is less than one, its return variations and expected return will be lower.
The expected return on a securities portfolio equals the risk-free rate of interest plus a risk premium, which compensates for the systematic risk the investor chooses to assume. Figure 2.16 is based on the assumptions that the risk-free rate of interest is 3 pct. and that the risk premium in the stock market is 3 pct. Consequently, the expected return on a portfolio with a beta of 1 will be 6 pct.
The relationship between beta and expected return is linear. An investor who divides his or her assets equally between a risk-free investment (with a beta of 0) and a portfolio of risky investments with a beta of 1, will have a beta for the overall portfolio that equals 0.5. Consequently, the portfolio is half as risky as the benchmark portfolio. The expected return on this portfolio will be equal to 4.5 pct., which is midway between 3 and 6 pct., cf. Figure 2.16. The price of reducing the risk is reflected in the 1.5-percentage point reduction in the expected return, compared to the expected return on the benchmark portfolio.
Figure 2.16 The relationship between the expected return, in percent, and the beta of the portfolio
Source Ministry of Finance
It is demonstrated, in the discussion of performance developments on the part of Folketrygdfondet, that the beta of the Government Pension Fund – Norway, relative to the benchmark portfolio, has been 0.88 over the period 1998–2006. Reduced risk implies a lower expected return. One may separate, for purposes of assessing the performance realised through the management effort, the positive or negative excess return achieved relative to the benchmark portfolio into two parts: One part that reflects the manager having adopted a different risk level than that of the benchmark portfolio (i.e. a beta of more or less than 1), and one part that may be attributed to the manager having achieved positive or negative excess return relative to the expected return pertaining to the relevant risk level. The latter component is termed alpha in financial literature.
Folketrygdfondet has delivered a negative excess return over the period 1998–2006, but when taking into consideration the low beta it turns out that the risk-adjusted excess return (“alpha”) has been positive. In terms of the Chart, this means that the return on the Government Pension Fund – Norway is on the upper side of the straight line, at the point where beta equals 0.88.
The beta of the Government Pension Fund – Global is calculated as 1.018, and is consequently much closer to 1 than that of the Government Pension Fund – Norway. In this simple example, the expected return on a portfolio with a beta of 1.018 would have been 6.05 pct., which is marginally higher than the expected return on the benchmark portfolio.
1 Systematic risk is discussed in-depth in financial literature. It has over time become common to relate systematic risk to several risk factors (not only to market beta). Consequently, the type of analyses presented here may be broadened. The illustration used here is based on the assumptions underpinning the Capital Asset Pricing Model (CAPM). The purpose of the example is only to illustrate interrelationships.
When the beta exceeds 1, it means that the return on the actual portfolio will on average be higher than the return on the benchmark portfolio during periods of positive returns in the securities market. In periods of negative returns, one would expect that the return on the Fund would on average be even weaker. 8 This may be illustrated by a numerical example: Assume that the benchmark portfolio registers a return of 5 pct one year. The Fund may then be expected to register a return of 5.09 pct., i.e. a excess return of 0.09 percentage point. 9 However, if the return on the benchmark portfolio was minus 5 pct., one would expect a 5.09 pct. decline in the value of the Fund, which represents a negative excess return of 0.09 percentage point.
In the analysis of the return on the actual portfolio, one has adopted the return on the benchmark portfolio as the only explanatory factor. However, financial literature argues that the difference between the returns on various securities and various securities portfolios may be attributed to more than one factor. For example, empirical studies of the stock market indicate that the return depends on the portfolio’s exposure to small companies, as well as its exposure to so-called “growth companies”, in addition to its exposure to the market in general (in the above analysis this is expressed through the benchmark portfolio). The Ministry will be refining the analyses with a view to including more explanatory factors.
The beta exposure has varied over time, cf. Figure 2.17. The data set indicates that the average beta over the last three-year period has been higher than during the initial part of the Fund’s history.
Figure 2.17 Beta developments for the Government Pension Fund – Global. Rolling twelve-month calculations. 1998–2006.
Source Norges Bank and the Ministry of Finance
II.4 Developments in the difference portfolio of the Government Pension Fund – Global
As discussed, Norges Bank may deviate from the benchmark portfolio in its management of the Government Pension Fund – Global, within a risk limit. The excess return provides a measure for the magnitude of the contribution to the overall return on the Fund that originates from Norges Bank’s active management effort. Table 2.5 shows the average excess return and the standard deviation of the excess return (actual tracking error) over the period 1998–2006, over the last three years and over the last year.
The information ratio, as set out in Table 2.5, is calculated as the ratio between the average excess return and the tracking error. The information ratio is in some contexts used as an indicator of the quality of active management. The ratio expresses how large the excess return has been for each unit of risk involved. The information ratio is used for comparing different managers, and for assessing developments over time for the same manager. The information ratio of the Government Pension Fund – Global is high compared to that of other funds, cf. Box 2.4.
Table 2.5 The excess return on and tracking error of the Government Pension Fund – Global, measured nominally in local currency. Percentage points and percentages1
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| The overall portfolio | |||
| Average excess return per year | 0.46 | 0.55 | 0.15 |
| Standard deviation | 0.38 | 0.34 | 0.37 |
| The information ratio | 1.2 | 1.6 | 0.4 |
| The equity portfolio | |||
| Average excess return per year | 0.70 | 0.87 | -0.01 |
| Standard deviation | 0.89 | 0.81 | 0.90 |
| The information ratio | 0.79 | 1.07 | -0.01 |
| The fixed-income portfolio | |||
| Average excess return per year | 0.22 | 0.31 | 0.24 |
| Standard deviation | 0.24 | 0.11 | 0.09 |
| The information ratio | 0.91 | 2.80 | 2.64 |
1 The annual excess return is calculated as the difference between the average monthly return on the Fund and on the benchmark portfolio, which is annualised by multiplying it by 12. This is given in percentage points in the Table. Actual tracking error is calculated as the standard deviation of the monthly excess return, which is annualised by multiplying it with the square root of 12. This is given in percentages in the Table.
Source Source: Norges Bank and the Ministry of Finance
The information ratio needs to be interpreted with care. A manager with a portfolio that does not deviate much from the benchmark portfolio will register low relative risk, and even a small excess return may result in a high information ratio. It is also the case that one would expect the information ratio to vary across different types of active management. An investor should primarily be concerned with maximising the excess return in NOK or in pct. at a given risk level, and not with maximising the information ratio.
Norges Bank has achieved excess return both for the period as a whole and for each individual year of the period. The excess return was less in 2006 than in the preceding periods. The average excess return over the period from 1998 is, in statistical terms, significantly higher than nil. Consequently, there is a low probability that this performance was achieved as the result of pure coincidence or luck. The accumulated gross excess return is about NOK 29 billion. 10
Alpha provides a measure of the excess return, as corrected for the beta exposure. For the Fund, alpha is estimated to be higher than nil, cf. Box 2.8. The difference between the excess return and alpha is calculated by multiplying the beta deviation (from 1) with the average return on the benchmark portfolio over the period. The intuition would be to expect a higher return on a portfolio with a beta in excess of 1, as compared to a benchmark portfolio (which by definition will have a beta of 1). Computations of alpha compare two portfolios; the return on the actual portfolio and the return on a “beta-adjusted” benchmark portfolio. Incidentally, the alpha of the Government Pension Fund – Global is, in statistical terms, significantly higher than nil.
Developments in actual tracking error are shown in Figure 2.18. Over the period from January 2000 and until December 2005, the Fund had a tracking error of about 0.25 pct. During the course of 2006, the tracking error increased somewhat, and reached 0.5 pct for a brief period in the summer of 2006. For the period as a whole, actual tracking error is calculated as 0.38 pct. Part of this risk may be attributed to a beta that differs somewhat from 1, cf. the above discussion. The contribution made by the beta deviation to the relative risk of the Fund is about 6 pct.
Figure 2.18 Developments in actual tracking error. Rolling twelve-month standard deviation of the excess return. 1998–2006. Percent.
Source Norges Bank and the Ministry of Finance
The expected tracking error shall not exceed 1.5 pct., cf. the limit stipulated by the Ministry of Finance. Thus far, Norges Bank has exploited less than one third of the allocated limit. This has to do with the orientation of the active management effort of Norges Bank, which is characterised by a large number of positions relative to the benchmark portfolio, and by the returns on such positions being mutually independent, both at a given point of time and over time. This type of active management strategy reaps major diversification benefits, thus implying a fairly low level of tracking error.
The actual portfolio of the Government Pension Fund – Global may be analysed as comprising two portfolios; the benchmark portfolio and the difference portfolio. Consequently, the risk of the Fund will depend on the risk of the benchmark portfolio, the risk of the difference portfolio (tracking error) and the covariation between the return on the benchmark portfolio and that on the difference portfolio.
The small difference between the absolute risk of the Fund and that of the benchmark portfolio as presented in Figure 2.14 is caused, inter alia , by a low correlation between the excess return and the return on the benchmark portfolio. For the period 1998–2006, this correlation coefficient is calculated as 0.25, cf. Box 2.1.
Charts 2.19 and 2.20 summarise parts of the above discussion. Figure 2.19 shows that the excess return for the period 1998 – 2006 has represented about 8 pct. of the overall return on the Fund. This illustrates the observation that the strategy of the Fund determines between 90 and 95 pct. of the return on the portfolio. For 2006, the excess return accounted for about 2 pct. of the overall return on the Fund. Figure 2.20 shows that the risk associated with active management resulted in an increase of about 2 pct. (2/98ths) in the overall risk of the Fund over the period 1998 – 2006 as a whole. For 2006 alone, the corresponding figure was 4 pct. Consequently, Norges Bank’s active management has made a significant positive contribution to the return on the Fund, without much impact on the risk of the Fund.
Figure 2.19 Overall return contribution from the benchmark portfolio and from the difference portfolio. 1998–2006. Percent.
Source Norges Bank and the Ministry of Finance
Figure 2.20 Overall risk contribution from the benchmark portfolio and the difference portfolio. 1998–2006. Percent.
Source Norges Bank and the Ministry of Finance
Table 2.5 also shows the excess return from equity and fixed-income management, respectively. Equity management has delivered a positive excess return over the period 1998–2006 and over the last three years, whilst there was a negative excess return for 2006. Tracking error has been calculated at about 0.90 pct., and has remained fairly stable over the various periods. Active fixed-income management has also delivered a positive excess return, when looking at the period as a whole. Fixed-income management achieved a positive excess return in 2006, which also resulted in a positive excess return for the Fund as a whole for that year. The relative risk associated with the fixed-income portfolio has been calculated as 0.24 pct. for the period as a whole, whilst the risk during the period since 2000 has remained stable, in the region of 0.10 pct. Higher risk during the first part of the period can mainly be attributed to two subsequent months in 1998, when the excess return was very high and very low, respectively. The information ratio associated with fixed-income management has been very high in recent years.
Active equity management has proven to be significantly more risky than active fixed-income management. Calculations show that the relative risk associated with equity management accounts for close to 90 pct. of the overall relative risk of the Fund. In other words, the variations in the Fund’s excess return are mainly driven by the variations in the equity portfolio’s excess return.
The Fund’s overall tracking error for the period as a whole has been 0.38 pct. This is significantly less than the weighted average of the relative risk of the equity portfolio and the fixed-income portfolio, respectively, which is about 0.5 pct. The explanation for this is low covariation between the rates of excess return from equity and fixed-income management, respectively. The correlation has been negative for the period as a whole. This is primarily caused by developments during the first two years. Since then, the correlation has on average been nil.
When using the return on the benchmark portfolio to assess the outcomes from Norges Bank’s active management, one has to be aware that one would not necessarily have managed to achieve the return on the benchmark portfolio by just managing the Fund close to index, cf. Box 2.1. Pure index management would incur transaction costs whenever the composition of securities in the benchmark portfolio changed, when there was an inflow of capital to the Fund and when there were changes to the Fund’s investment strategy. At the same time, index management may generate relatively high lending income from securities in the portfolio. Such lending income may in large part cancel out the regular transaction costs associated with index management.
II.5 Developments in the benchmark portfolio of the Government Pension Fund – Norway
The Ministry has stipulated general investment limits for the Government Pension Fund – Norway, whilst the more detailed benchmark portfolio for measurement of return and risk is defined by the Executive Board of Folketrygdfondet. The return on the Government Pension Fund – Norway, which is primarily invested in the Norwegian securities market, depends in large part of market developments in the benchmark portfolio. The benchmark portfolio for the equity investments in Norway is the main index of the Oslo Stock Exchange, whilst FTSE NOREX 30 has been adopted as the benchmark portfolio for the equity investments in Denmark, Finland and Sweden. 11 The fixed-income benchmark portfolio has been a composite index reflecting developments in the Norwegian fixed-income market.
Returns and risks associated with the overall benchmark and the sub-benchmarks for various periods are set out in Table 2.6. The return on the overall benchmark portfolio over the last three-year period has been significantly higher than the average return for the period 1998–2006 as a whole. The risk, as measured by the standard deviation of the return, has been somewhat higher during the last part of the period than for the period as a whole.
Table 2.6 Average annual return and annual standard deviation of the return on the benchmark portfolio of the Government Pension Fund – Norway, measured nominally in Norwegian kroner. Percent1
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| The benchmark portfolio | |||
| Average return per year | 7.38 | 10.81 | 9.64 |
| Standard deviation | 4.71 | 5.09 | 4.33 |
| Equity benchmark – Norway | |||
| Average return per year | 12.02 | 33.29 | 29.56 |
| Standard deviation | 22.49 | 16.67 | 15.83 |
| Equity benchmark – the Nordic region | |||
| Average return per year | 8.85 | 22.06 | 23.98 |
| Standard deviation | 23.54 | 14.96 | 16.21 |
| The fixed-income benchmark | |||
| Average return per year | 5.81 | 3.80 | 0.74 |
| Standard deviation | 3.05 | 3.13 | 1.88 |
1 For the equity investments in the Nordic region, the time period is June 2001-2006.
Source Folketrygdfondet and the Ministry of Finance
Returns in the Norwegian stock market have been considerably higher than returns in the Nordic stock market, as measured by developments in the two indices. However, risk has been fairly similar in the two markets over this period. Since 2004, returns in the stock markets in Norway and the Nordic region have been very high.
Returns in the Norwegian bond market, as measured by the return on the fixed-income benchmark of the Fund, have been favourable when looking at the period as a whole, but were very low in 2006. The risk associated with investments in the Norwegian bond market was also low in 2006.
The overall benchmark portfolio of the Government Pension Fund – Norway is a composite of the abovementioned benchmark sub-portfolios. As per the end of November 2006, the benchmark portfolio for investments in Norwegian equities carried a weight of close to 23 pct. in the overall portfolio, the benchmark portfolio for investments in Nordic equities carried a weight of 4 pct., whilst the fixed-income benchmark carried a weight of 73 pct. In December 2006, the sight deposits with the treasury were terminated, and the equity portfolio therefore represents a larger portion of the overall portfolio than was previously the case. Consequently, one should be prepared for larger annual fluctuations in the return on the Pension Fund – Norway in future than have been registered historically. Some simple calculations indicate that the writedown of the sight deposits results in the risk associated with the Fund being more or less doubled, and that the standard deviation of the overall portfolio may be expected to increase from about 5 to 10 pct.
II.6 Developments in the actual portfolio of the Government Pension Fund – Norway
The benchmark portfolio largely determines how the capital is invested. The deviation between the actual portfolio and the benchmark portfolio can mainly be attributed to the investment decisions of Folketrygdfondet. The return on, and the risk associated with, the actual portfolio is shown in Table 2.7. The average return on the actual portfolio has been less than that on the benchmark portfolio, with the exception of 2006. For 2006, the opposite is the case. As shown in Table 2.7, the risk associated with the actual portfolio has been somewhat less than the risk associated with the benchmark portfolio, with the exception of 2006. This is also shown in Charts 2.21 and 2.22.
Figure 2.21 Return on the Government Pension Fund – Norway and on the Fund’s benchmark portfolio. Monthly return data 1998–2006. Percent
Source Folketrygdfondet and the Ministry of Finance
Figure 2.22 Rolling twelve-month standard deviation of the return on the Government Pension Fund – Norway and on the Fund’s benchmark portfolio. 1998–2006. Percent
Source Folketrygdfondet and the Ministry of Finance
Table 2.7 Average annual return and annual standard deviation of the return on the Government Pension Fund – Norway, measured nominally in Norwegian kroner. Percent.
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| Actual portfolio | |||
| Average return per year | 7.29 | 9.90 | 10.68 |
| Standard deviation | 4.26 | 4.60 | 4.92 |
Source Folketrygdfondet and the Ministry of Finance
The excess return on the Fund, i.e. the difference between the return on the actual portfolio and on the benchmark portfolio, is discussed in II.7 for the various sub-portfolios.
The Ministry’s calculations show that the variations in the return on the benchmark portfolio explain about 93 pct. of the variations in the Fund’s return. This illustrates that the Ministry’s general investment strategy, as expressed through the investment limits laid down in the Regulations governing the management of the Government Pension Fund – Norway and as operationalised in the benchmark portfolio selected for the Fund by the Executive Board, largely determine developments in the return on the actual portfolio.
In Figure 2.23, the information in Figure 2.21 is organised somewhat differently. Each dot shows the return on the benchmark portfolio and on the actual portfolio for the same month, as sorted by the level of return. The dots are clustered closely along a straight line, as drawn in the Chart. The slope of this line, which is termed the “beta” of the portfolio, is calculated to be 0.875, which in statistical terms is significantly less than 1, cf. Box 2.8. 12 When the beta is less than 1, it means that the return on the actual portfolio will on average be less than the return on the benchmark portfolio during periods of positive market developments. 13 The Norwegian equity portfolio has featured a beta of about 0.92 for this period, whilst it has been about 1 for the Nordic equity investments. Consequently, this reaffirms the observation that the Norwegian equity portfolio of Folketrygdfondet tends to outperform the benchmark portfolio during periods of weak developments on the Oslo Stock Exchange, and to somewhat underperform the benchmark portfolio during periods characterised by a particularly strong upturn in equity prices. Folketrygdfondet is a large player in the Norwegian stock market. The portfolio of the Government Pension Fund – Norway is concentrated on a smaller number of companies than those included in the benchmark portfolio (the main index of the Oslo Stock Exchange). The main reason why the beta is less than 1 is that Folketrygdfondet has opted to refrain from investing in certain companies carrying a particularly high risk relative to the average risk of the benchmark portfolio.
Figure 2.23 Correlation between the return on the Government Pension Fund – Norway and on the Fund’s benchmark portfolio. 1998–2006. Percent.
Source Folketrygdfondet and the Ministry of Finance
Figure 2.24 shows developments in the portfolio’s beta over time. As may be seen from the Chart, the beta has been less than 1 for long periods of time. This means that the systematic risk associated with the benchmark portfolio has been higher than the risk associated with the actual portfolio of the Government Pension Fund – Norway.
Figure 2.24 The Government Pension Fund – Norway. Rolling twelve-month calculations. 1999-2006.
Source Folketrygdfondet and the Ministry of Finance
The fluctuations in the portfolio’s beta during 2000–2002 need to be analysed in the context of the steep decline in the equity prices of technology companies, which impacted considerably less on the actual equity portfolio of Folketrygdfondets than on the benchmark portfolio. The lower beta during 2003–2005 has to do with, inter alia , Folketrygdfondet reducing the duration of its fixed-income portfolio compared to that of the benchmark portfolio, cf. Box 3.2. Throughout 2006, the difference in duration between the fixed-income portfolio and the benchmark portfolio has been reduced. At the same time, Folketrygdfondet has featured a higher equity portion than suggested by the Fund’s benchmark portfolio. Both these factors have contributed to increasing the beta of the portfolio.
II.7 Developments in the difference portfolio of the Government Pension Fund – Norway
In its management of the Government Pension Fund – Norway, Folketrygdfondet may, within a risk limit, deviate from the benchmark portfolio. The contribution from such active management is measured on an ongoing basis, by way of developments in the value of the Fund being compared to developments in the benchmark portfolio.
Table 2.8 shows average excess returns, the standard deviation of excess returns (actual tracking error) and the information ratio over the periods 1998–2006, the last 3 years and the last year, in respect of the Fund as a whole, the equity portfolios of Norway and the Nordic region, respectively, as well as the fixed-income portfolio.
Table 2.8 Excess return and tracking error of the Government Pension Fund – Norway, measured nominally in Norwegian kroner. Percentage points and percent1
| 1998 – 2006 | 2004 – 2006 | 2006 | |
|---|---|---|---|
| The overall portfolio | |||
| Average excess return per year | -0.09 | -0.91 | 1.04 |
| Standard deviation | 1.25 | 1.25 | 0.83 |
| The information ratio | -0.1 | -0.7 | 1.3 |
| The Norwegian equity portfolio2 | |||
| Average excess return per year | 0.69 | -2.96 | -0.44 |
| Standard deviation | 4.68 | 3.93 | 4.22 |
| The information ratio | 0.15 | -0.75 | -0.10 |
| The Nordic equity portfolio3 | |||
| Average excess return per year | -0.40 | -0.36 | -0.97 |
| Standard deviation | 0.54 | 0.55 | 0.62 |
| The information ratio | -0.74 | -0.65 | -1.57 |
| The fixed-income portfolio | |||
| Average excess return per year | -0.34 | -0.75 | 0.59 |
| Standard deviation | 0.83 | 1.20 | 0.42 |
| The information ratio | -0.41 | -0.62 | 1.40 |
1 The annual excess return is calculated as the difference between the average monthly return on the Fund and on the benchmark portfolio, and is annualised by multiplying by 12. It is reported in percentage points in the Table. The actual tracking error is calculated as the standard deviation of the monthly excess return, and is annualised by multiplying by the square root of 12. It is reported in pct. in the Table.
2 In 2006, the Norwegian equity portfolio included equities in the company General Electric, which Folketrygdfondet received in settlement in connection with the sale of Nycomed Amersham. The value of these equities underperformed relative to the main index of the Oslo Stock Exchange last year, thus contributing to the negative excess return.
3 The figures for the Nordic equity portfolio cover the period from May 2001 until the end of 2006.
Source Folketrygdfondet and the Ministry of Finance
The excess return attributable to Folketrygdfondet was negative over the period 1998–2006. Performance over the last three-year period was weaker than for the period as a whole. However, this changed in 2006, and the Fund achieved a positive excess return as the result of, inter alia , good performance as far as fixed-income management was concerned.
The excess return on the Fund deviates from a weighted average of the excess return on the three sub-portfolios. This can be attributed to tactical allocation decisions, i.e. that Folketrygdfondet has at times been over- or underweighted in the various asset classes (Norwegian equities, Nordic equities, Norwegian fixed-income securities) relative to the asset class weights featured by the overall benchmark portfolio.
Alpha provides a measure of the excess return as adjusted for the low beta exposure of Folketrygdfondet. Alpha is estimated to be higher than nil, cf. Box 2.8. The reason why alpha is positive, despite the excess return having been negative, is that this calculation “compensates” the return on the actual portfolio for a beta of less than 1. Sub-chapter II.4 contains a more detailed discussion of how caution needs to be exercised in interpreting the information ratio, and how it may vary across different styles of active management.
The Government Pension Fund – Norway adopts a long investment horizon. The Fund is a long-term investor, and will maintain a large portfolio of equities in the Norwegian market. In its investment strategy, Folketrygdfondet shall emphasise the long-term growth potential of the companies, based on an assessment of fundamental factors and developments. Consequently, short-term market fluctuations are of less importance. In line with this perspective, the Ministry focuses on developments in the Pension Fund over time in its follow-up of Folketrygdfondet’s performance. Over the period 1998–2006, Folketrygdfondet has achieved a return that is more or less in line with the return on the benchmark portfolio. When taken in isolation, Folketrygdfondet has achieved a positive excess return in its management of Norwegian equities over this period.
Since Folketrygdfondet began investing in Denmark, Finland and Sweden in 2001, its management of the Nordic equity portfolio has, in aggregate over these years, contributed a negative excess return. Some of the negative excess return from this part of the portfolio reflects the fact that Folketrygdfondet pays withholding tax on equity dividends. The Ministry is aware that Folketrygdfondet has made changes to its operational management arrangements as far as Nordic equities are concerned, with a view to improving performance. Fixed-income management performed well in 2006. This was because the portfolio assumed less interest rate risk (as measured by duration, cf. Box 3.2) than did the benchmark portfolio, during a period when market interest rates were on the increase. The weaker performance in the preceding years was caused by the portfolio having less interest rate exposure during a period of declining market interest rates.
The difference in return between the actual portfolio and the benchmark portfolio has been most pronounced for the asset class Norwegian equities. It follows from Table 2.8 that this asset class has also exhibited the highest tracking error by far.
Developments in the actual tracking error of the overall portfolio are shown in Figure 2.25. For the period as a whole, actual tracking error is calculated at 1.25 pct., which is somewhat lower than the upper limit of 2.0 pct. defined by the Executive Board for the overall portfolio. Tracking error declined somewhat during the course of 2006, and was about 0.8 pct towards the end of the year. For the period as a whole, actual tracking error is calculated at 1.25 pct. Much of this risk may be attributed to a beta that deviates somewhat from 1, cf. the above discussion over. The beta deviation accounts for about 22 pct. of the Fund’s relative risk.
Figure 2.25 Developments in actual tracking error. Rolling twelve-month standard deviation of the excess return. 1998–2006. Percent.
Source Folketrygdfondet and the Ministry of Finance
The Fund’s overall tracking error for the period as a whole is less than the weighted average of the tracking error of the equity and fixed-income portfolios, which may under certain simplified assumptions be estimated at 1.7 pct. This is explained by low covariation between the rates of excess return from equity and fixed-income management, respectively.
Footnotes
The value of the Government Pension Fund – Global is calculated before the deduction of management costs. In 2006, management costs amounted to about NOK 1.5 billion, which will be reimbursed to Norges Bank during the course of the 1st quarter 2007.
The estimated increases in the value of the Government Pension Fund and the Government Pension Fund – Global in Sub-chapter 2.1 are based on figures from the Central Government Accounts up to and including 2005 and from Norges Bank’s annual report for 2006. See also Box 1.1.
The real return is approximately the same as the nominal return less inflation.
“Economies of Scale, Lack of Skill or Misalignment of Interest? A Study on Pension and Mutual Fund Performance.” (Working paper that will be published)
Folketrygdfondet had liabilities of NOK 2.2 billion in relation to, inter alia , repo trades drawn from the fixed-income investments of Folketrygdfondet, cf. Folketrygdfondet’s annual report on the management of the Government Pension Fund – Norway in 2006.
Folketrygdfondet has managed a Nordic equity portfolio since 2001. Parts of the negative excess return on such portfolio needs to be seen in the context of withholding tax paid on equity dividend over the period 2001-2004.
Standard statistical methods are adopted for purposes of calculating the alpha and beta.
The statement is strictly correct if the intercept (alpha) equals zero for purposes of the regression analysis.
5.09 pct. corresponds to 5 pct. multiplied by a beta of 1.018.
The term gross excess return does not reflect, inter alia , transaction and management costs associated with index management, or additional management costs associated with active management.
The index comprises the 30 largest and most liquid companies in the Nordic region. The Norwegian companies encompassed by the index have been excluded for purposes of composing the Fund’s benchmark portfolio.
Standard statistical methods have been adopted in calculating alpha and beta.
The statement is strictly correct if the intercept (alpha) equals zero for purposes of the regression analysis.