3 Government Pension Fund Global: Further development of strategy and management

3.1 Geographical distribution of the equity benchmark

3.1.1 Background

In the Government Pension Fund 2019 white paper, the Ministry of Finance announced a review of the framework and benchmark for GPFG equity investments, including the geographical distribution of the benchmark index. The Ministry stated that the equity benchmark shall support important considerations underpinning the investment strategy, such as broad diversification of risk, harvesting of risk premiums and cost efficiency. It was further recognised that regular reviews serve to ensure that the composition of the benchmark index is well adapted to the objective and distinctive characteristics of the Fund, whilst also reflecting new knowledge and equity market developments. The Ministry intended to present its analysis and recommendations in the Government Pension Fund 2020 white paper.

In a letter of 6 November 2018 from the Ministry of Finance to Norges Bank, the Bank was asked for its advice and assessments on the composition of the equity benchmark. Norges Bank was also asked to provide an account of its experience investing in emerging markets and markets not included in the benchmark index, and to describe its framework for managing and controlling the distinctive risks associated with these markets.

Norges Bank submitted its analysis and advice to the Ministry in letters of 21 and 22 August 2019, respectively. As part of its basis of assessment, the Ministry has furthermore commissioned a report from consultancy firm and index provider MSCI on global equity market developments and analysis of index risk and return for various geographical distributions. The letters from Norges Bank and the MSCI report are available on the Ministry website.

Based on an overall assessment, the Ministry is proposing to modify the GPFG equity benchmark adjustment factors resulting in the share of developed markets in Europe being reduced somewhat, and the North American share correspondingly increased. The proposed change is in line with the recommendation of Norges Bank. The Ministry considers the modification of the benchmark index to provide a better representation of economic activity in listed companies globally.

The current benchmark index is detailed in section 3.1.2. The analysis and recommendations of Norges Bank are discussed in section 3.1.3. The MSCI report is described in section 3.1.4, whilst the Ministry’s assessments and recommendations are presented in section 3.1.5. This chapter is supplemented by three indepth discussion articles (in Norwegian only); chapter 7 analyses the implications of the chosen regional distribution in terms of historical risk and return, chapter 8 provides a detailed presentation of certain aspects of the analysis in MSCI’s report to the Ministry, and chapter 9 discusses academic studies on the financial market impact of past pandemics.

3.1.2 The current benchmark index

The GPFG investment strategy is reflected in a strategic benchmark index stipulated by the Ministry of Finance. The benchmark index comprises an equity share of 70 percent and a fixed-income share of 30 percent; see chapter 2.1 (in Norwegian only). Norges Bank may deviate from the benchmark index only within a set risk limit. The composition of the benchmark index is therefore of great importance for the actual investments, as well as for Fund risk and return.

Up until 2012, the equity benchmark had a fixed geographical distribution, with 50 percent in Europe, 35 percent in the Americas and 15 percent in Asia-Pacific. This geographical distribution was established with considerable emphasis on foreign exchange risk. The assessment was based on an assumption that a distribution between the geographical regions in the benchmark index matching the composition of Norwegian imports would reduce foreign exchange risk, thereby furthering the objective of maximising the international purchasing power of the fund capital. Since Europe is the origin of a considerable proportion of Norwegian imports, Europe accounted for a large portion of the benchmark index.

In 2011 and 2012, the Ministry presented an updated analysis on foreign exchange risk and the geographical distribution of the investments of the Fund. The review indicated that the foreign exchange risk in the GPFG appeared to be lower than previously assumed. The Ministry concluded that it was no longer appropriate to hold such a high share of the equity investments in Europe; see the discussion in the white papers on the Government Pension Fund published in 20111 and 20122. The Ministry suggested that free float-adjusted global market weights were an appropriate basis for the geographical distribution of the GPFG equity benchmark.

Against this background, the geographical distribution of the benchmark index was changed in 2012. The fixed geographical distribution was replaced by market weights. Consequently, the geographical distribution between various regions is no longer fixed, and will fluctuate with market developments. In addition, a distinction between developed and emerging markets was introduced.

The new geographical distribution from 2012 still involved a higher share in developed markets in Europe and a smaller share in developed markets in North America than what would be implied by free float-adjusted market weights. The share in other developed markets and emerging markets was more or less in line with free float-adjusted market weights. In the white paper on the Government Pension Fund published in 2012, the Ministry emphasised that a full adjustment to free float-adjusted market weights would entail halving the share invested in Europe, based on the then market values. The Ministry emphasised that this would be a considerable change, and that the resulting high concentration in developed markets in North America would make the benchmark index more vulnerable to developments in the US economy. The chapter 7 discussion article presents an analysis of the historical risk and return implications of the chosen regional distribution of the benchmark index.

The GPFG equity benchmark is based on the global equity index FTSE Global All Cap, from the index provider FTSE Russell. The FTSE Global All Cap index is intended to provide broad representation of the international equity market, and includes all markets classified by the index provider as developed, advanced emerging or secondary emerging; see table 3.1. The index encompasses large, medium-sized and small companies in the respective markets. Criteria for inclusion of securities in the index are set by the index provider. The methodology underpinning the index is rule-based and made publicly available on the FTSE Russell website.

The composition of the FTSE Global All Cap index is based on the total market capitalisation of the companies included in the index. The capitalisation of each company is subsequently adjusted, among other things, for liquidity, foreign ownership restrictions and free float.3 The purpose of such adjustments is to enable investors to closely replicate the index at low cost.

The Ministry has further assigned adjustment factors to the market capitalisation in the index, on basis of the regional affiliation of companies: developed markets in Europe, developed markets in North America, other developed markets and emerging markets. The adjustment factors affect ownership stakes in all companies in each region.

At year-end 2019, developed markets in Europe, North America, and other developed markets, accounted for 33 percent, 42 percent and 14 percent, respectively, of the equity benchmark for the Fund. Emerging markets accounted for about 11 percent; see figure 3.1 in section 3.1.4.

Emerging markets sub-index

Index providers like FTSE Russell and MSCI classify markets based on a number of criteria. A key distinction is made between developed and emerging markets. A market classified as emerging meets some, but not all, of the criteria required to qualify for developed market status.4

Table 3.1 shows FTSE Russell’s classification of global stock markets at year-end 2019. The index provider updates on an annual basis, a watch list of markets that meet, or are close to meeting, the criteria for inclusion in the FTSE Global All Cap index. Over time there have been changes in which markets are classified as emerging.5

The index provider FTSE Russell also classifies many markets as frontier markets. These markets are held to be less developed than markets that have been accorded emerging market status. Frontier markets are neither included in the FTSE Global All Cap index, nor in the equity benchmark for the Fund. Moreover, a considerable number of markets are unclassified by the index provider. Unclassified markets are not included in the GPFG benchmark index.

The assessments of the index provider, which underpin the classification of markets, are based on 21 criteria. These criteria include, among other things, how the markets function, regulatory matters, settlement, custodianship and trading conditions, as well as the extent to which the marketplace has a well-developed derivatives market.6 The criteria do not include considerations of political risk or risk relating to environmental, social or corporate governance issues.

The GPFG equity benchmark has over time been expanded to include additional emerging markets; from six in 2000 to 23 at year-end 2019.7 The benchmark index composition of emerging markets does, as a main rule, mirror that of the FTSE Global All Cap index. Hence, changes in which markets are included in the underlying index will normally also be reflected in the GPFG equity benchmark. Pending the Ministry’s review of the benchmark index, the Government Pension Fund 2019 white paper stated that new markets would not be included in the benchmark until a decision has been made on the composition thereof.

The mandate from the Ministry of Finance to Norges Bank requires the Executive Board to preapprove all markets in which the fund capital is to be invested. The Bank is also required to document adequate procedures for such approvals, see box 2.5 (in Norwegian only), which discusses an independent review of the Bank’s procedures, carried out by Norges Bank’s external auditor. The approval requirement applies irrespective of whether or not the relevant market is included in the benchmark index. The Executive Board has, with the exception of Pakistan, approved all of the markets included in the benchmark index. In addition, the Bank has approved 26 markets outside the benchmark. These include both frontier markets and unclassified markets.

3.1.3 Norges Bank’s advice and assessments

Regional distribution

In a letter of 21 August 2019, Norges Bank recommended to further adjust the geographical distribution of the benchmark index towards free float-adjusted market weights by increasing the share in North America and reducing the share in European developed markets. No changes were proposed for the share in emerging markets or other developed markets.

In its letter, the Bank notes that the investment strategy for the GPFG implies that risk and return predominantly are determined by the benchmark index, and that the benchmark therefore is of crucial importance to the investment management. The Bank states that the index should be composed on the basis of transparent, verifiable criteria, and be investable in order to serve as a long-term management performance benchmark. The Bank is of the view that the free float-adjusted market capitalisation index FTSE Global All Cap meets these criteria. Any deviations from such an index should be justified and serve a specific purpose.

Norges Bank states that distinctive characteristics of the Fund may serve to justify a geographical allocation that deviates somewhat from free float-adjusted market weights, but is nonetheless of the view that the geographical distribution should be further modified in the direction of free float-adjusted market weights.8 One such distinctive characteristic of the GPFG is the long investment horizon. According to the Bank, the relevant risk for long-term investors is the risk of permanent loss. A permanent loss may arise as the result of changes in expected corporate earnings. If events affect expected earnings in a market or a sector with a high benchmark weighting, the permanent loss may be large. The Bank however emphasises that it is difficult to assess the likelihood of such events and the extent to which such risk is reflected in the market prices.

Norges Bank also comments that the low probability of unexpected withdrawals makes the Fund well suited to carry the risk and harvest the potential gains from investing in less tradable assets. Free float-adjustments on part of the index provider to the market capitalisation of individual companies are made to enable investors with continual liquidity needs to closely replicate the index. However, the GPFG is not such an investor. A free float-adjusted index represents about 80 percent of the underlying market capitalisation of the companies. The Bank further observes that 55 percent of global economic activity in 2018 took place in emerging and frontier markets. In comparison, these markets represented only just above 20 percent of the global listed equity market. The Bank states that the Fund may be well suited to invest in less liquid emerging markets, and thereby achieve a better diversified portfolio.

Finally, Norges Bank comments that in determining the geographical distribution, one in principle might attach weight to the Fund’s position in individual markets, for example regarding tax and corporate governance issues, given that the Fund in selected markets has a position that is permanently and materially different from that of the marginal investor. The Bank however recognises that this is not easy to establish, and states that tax rules in any given market may vary with the type of investor and with the value and time horizon of the investments. In addition, tax rules and corporate governance practices will change over time. The Bank observes that minority shareholder rights appear to be better protected in Europe, but also that there are large differences between countries within Europe. According to the Bank, there is reason to assume that general differences in corporate governance practices between countries are reflected in market prices.

In a letter of 29 April 2020, the Ministry of Finance asked Norges Bank to examine whether any considerations relating to the current coronavirus outbreak might potentially merit modification of the Bank’s advice. In its reply of 29 May 2020, the Bank states that the coronavirus outbreak has triggered major market fluctuations and that significant return differences have been observed across both sectors and countries. These recent market development highlight, in the view of the Bank, the importance of being broadly diversified across sectors and markets. Norges Bank again emphasises that the relevant risk for long-term investors is the risk of permanent loss. The Bank states that even if one were to assume that the risk of permanent loss has increased as a result of the coronavirus outbreak, this will only be of relevance to the issue of geographical distribution if these losses affect the various regions differently, and only if the risk of this happening is not reflected in market prices at the time of making the modifications. The Bank observes that such assumptions currently do not seem to apply, and therefore sees no need for changing the advice set out in the letters of 21 and 22 August 2019.

Emerging markets

In its letter of 22 August 2020, Norges Bank does not recommend any changes to the method and rules for the composition of the emerging markets sub-benchmark. The Bank believes that the benchmark index should, like at present, include all companies in all developed and emerging equity markets that are included in the FTSE Global All Cap index.

In its letter, the Bank states that the «(…) experience from almost 20 years as a state owned minority shareholder in emerging market companies is positive». The Bank’s framework for risk management and control in emerging equity markets has evolved over time and does, according to the Bank, provide for comprehensive assessment of both the financial and non-financial risks the fund is exposed to through such investments. The Bank believes that the current framework supports the investment objective of achieving the highest possible return, given an acceptable level of risk.

Norges Bank further explains how the framework for risk management and control has been implemented in accordance with the requirements under the management mandate. The mandate requires the Bank to establish principles for the management, measurement and control of risk, as a minimum adhering to internationally recognised standards and methods. The Executive Board has adopted supplementary risk management guidelines for the Bank. These address various types of risk, such as strategic, investment-related and operational risk, and how these risks shall be managed and controlled. The Bank notes that political risk may affect the value of the investments and in such cases fall within the scope of investment risk. Furthermore, the Bank highlights that the Executive Board requires the risk of reputational loss to be assessed. The NBIM Chief Executive Officer may accept, but is required to inform the Executive Board of, any operational risk or risk of reputational loss that is considered to be significant. Any operational or reputational risk that is held to be of critical importance may be accepted by the Executive Board only. These general principles form the basis for the Bank’s approval procedures of equity markets. Norges Bank manages the risk associated with individual investments in emerging stock markets through, inter alia, the use of locally based external managers and through risk-based divestments.

The Bank further states that constructing an emerging markets sub-index that both meets the requirements applicable to a well-designed benchmark, and at the same time with a high degree of certainty reduces exposure to political risk would be challenging. The most effective approach to reducing the financial implications of political risk would, generally speaking, be to diversify the investment across a large number of equity markets. This suggests, in the view of the Bank, that the benchmark index should be broad and that the Fund should, like at present, be permitted to invest in markets that are not included in the benchmark.

The Bank recognises that the significance of country-specific considerations in emerging markets may nonetheless suggest that one might want to stipulate a cap on the portion of the benchmark one single emerging market may be permitted to constitute. Special weight may be attached to such considerations if high exposure is accompanied by high risk, including non-financial risk. In a letter of 15 April 2020, the Ministry asked for the Bank’s assessment as to how a potential cap might be structured and what implications it would involve. The Ministry received the Bank’s assessments on 27 August 2020.

In its letter of 22 August 2019, Norges Bank did not discuss ethical considerations relating to investments in individual countries, but notes that the Ministry has appointed a committee to review the ethical guidelines for the GPFG. The committee submitted its report on 15 June 2020. The report is discussed in chapter 6.3 (in Norwegian only).

3.1.4 Report from MSCI

The consultancy firm and index provider MSCI has been commissioned by the Ministry of Finance to prepare a report describing developments in the global stock market and analysing the historical risk and return implications of different geographical distributions. The analysis has been carried out primarily on the basis of MSCI’s global indices. Chapter 8 provides a detailed presentation of selected parts of the analysis.

In its report, MSCI refers to the increased globalisation of the world economy and financial markets. A key development has been the increased importance of emerging economies. According to MSCI, sales revenues in emerging economies accounted for 40 percent of sales revenues in listed companies in 2019. New markets and emerging market companies are also being included in global equity indices. Investors are increasingly diversifying their equity investments globally; see box 3.1. MSCI comments that these developments have made market risk more complex and multifaceted.

Despite developments towards closer economic and financial integration, the analysis shows that there is still a significant diversification benefit from spreading investments globally. It would appear that such a diversification benefit has generally not been reduced in a free float-adjusted market-weighted index in the last 20 years, although it has declined during periods of global equity market turbulence.

The analysis shows that risk and return in broad, global portfolios will be systematically affected by a number of factors. Developments in regions, countries, industries and other systematic risk factors, such as variations in investor preferences for liquidity, are of significance. Broad diversification across all these dimensions may serve to reduce overall risk in the portfolio. Investors should, according to MSCI, also consider the impact of individual companies with particularly high market value, since variations in their return may influence performance figures for an entire region.

MSCI states that regions are meaningful building blocks in composing an investment portfolio. Economic integration is closer between countries within the same region, thus implying a higher correlation between markets within a geographical region than between markets in different regions. Emerging markets differ from developed markets because, inter alia, country-specific risk is more prominent. The analysis also shows that there are considerable benefits from diversifying investments across regions, and that such diversification benefits may be larger in the long run than in the short run. The correlation between regions has fluctuated over time and has increased somewhat over the last 20 years. MSCI’s analysis also shows that the correlation is reduced when the time period analysed is longer; see the chapter 8 discussion article.

MSCI finds that there have historically been large differences in long-term returns between regions. The return differences may be linked to differences in profitability and valuation levels of listed companies, as well as developments in these variables over time. Declining valuations of the profits of Japanese companies are, for example, an important cause as to the low average returns in Asian markets. In the time period examined by MSCI, North American companies stand out by having significantly higher profitability than companies in other regions.

MSCI has also analysed geographical distributions based on different index weighting principles, including full and free float-adjusted market values, region size as measured by GDP and equal weighting by region market values or risk; see chapter 8. MSCI finds that the various weighting principles give rise to substantial differences in geographical distribution, although differences in risk and diversification benefits have been very limited in the short run. Nor have differences in the liquidity of the indices had much impact on historical returns. A free float-adjusted index does, however, stand out by resulting in higher country concentration than do other weighting principles. MSCI states that long-term investors should consider country concentration, and that the risk associated with such concentration is not adequately captured by short-term risk measures. A high country concentration increases the vulnerability of investments to developments in individual regions, and reduces diversification of the portfolio’s exposure to regional differences in long-term company performance.

MSCI has also examined whether some regions offer higher expected return than others. The analysis indicates that emerging markets entail higher market risk, lower market values, lower liquidity, and higher political risk than developed markets. Such characteristics have, according to MSCI, served to explain return differences between markets globally. MSCI writes that the expected equity premium is likely to be higher in emerging markets and that investors are compensated for the higher risk involved. MSCI has also examined whether differences in economic growth between countries can explain long-term return differences between countries, but finds no strong correlation.

3.1.5 The Ministry’s assessments

Key considerations

The GPFG investment strategy is based on the premise that risk can be reduced by diversifying investments across various asset classes, countries, sectors and individual companies. Future developments in the world economy are uncertain and will vary between countries and regions. Moreover, the conditions under which listed companies and investors operate in individual countries are influenced by systems of government as well as political developments. A broad geographical diversification of the investments makes the Fund less vulnerable to developments in individual countries, smoothens portfolio fluctuations and serves to improve the ratio between expected return and financial risk.

The investment objective for the GPFG is to achieve the highest possible return measured in international currency, given an acceptable level of risk. When assessing the equity benchmark, the Ministry emphasises cost effectiveness, broad diversification of the investments and harvesting of risk premiums. The benchmark index shall be rule-based, verifiable and transparent, as well as adapted to the distinctive characteristics of the Fund, such as size and investment horizon.

Market weights and adjustment factors

Investing in the equity market implies holding ownership stakes in listed companies. An index based on the market capitalisation of companies offers a broad representation of economic activity in listed companies globally as well as broad diversification of investments across countries, industries and companies. The weight attributed to each country is determined by its listed companies’ share of the global stock market, measured by market value. Hence, the geographical distribution of such an index reflects the capital distribution across markets and developments therein.

Market capitalisation is also the theoretically preferred principle for the composition of a global equity index. An investor tracking a market-weighted index will hold equal ownership stakes in all companies included in the index, mirroring the investments of the average investor and delivering the same risk and return as the global equity market. If an investor opts for a different distribution, it may be because such an investor is faced with other investment conditions, assesses the investment environment differently or has other investment objectives than the average investor. Box 3.1 discusses factors that may influence the distribution of global equity investments.

The composition of the current GPFG equity benchmark is based on full market capitalisation, but with some modifications. Firstly, the index provider adjusts the market capitalisation for, inter alia, liquidity, foreign ownership restrictions and free float. The free float-adjustment is the modification of most importance to investors. The starting point of a free float-adjusted index is company market values, but the index weight of a company is reduced when parts of its market value is not readily available for public trading. This typically applies in cases where long-term investors, such as families, entrepreneurs and governments, hold ownership stakes. The free float-adjustment thus provides a better measure of stock tradability and market investment capacity. The Ministry is of the view that a free float-adjusted index offers a sound basis for cost-effective management of a large fund like the GPFG. However, higher index liquidity may imply lower expected return if investors require compensation for low liquidity. The Ministry has on the other hand noted that MSCI finds liquidity differences between indices to provide little explanation for historical return differences.

Concurrently, the Ministry recognises that a free float-adjusted index reflects a smaller part of global market value and entails increased country concentration compared to the distribution of underlying market values. About 70 percent of global market value is covered in the free float-adjusted index, but the scale of adjustment varies significantly between regions; see figure 3.1. Developed markets in North America represent about 45 percent of total global market value, whilst North American developed markets constitute as much as 58 percent of the free float-adjusted index underpinning the GPFG equity benchmark.

Zoom in on image

Source FTSE Russell, Macrobond, MSCI and the Ministry of Finance.

The GPFG benchmark index has further been modified by assigning adjustment factors based on geographical affiliation to the free float-adjusted market capitalisation of companies. These adjustment factors are determined by the Ministry, and result in the Fund benchmark index having a higher share in developed markets in Europe, and a smaller share in North America (United States and Canada), than would be implied by both full and free float-adjusted market weights; see figure 3.1.

The adjustment factors in the GPFG benchmark index are applied on the assumption that regions can be used to manage the benchmark risk composition. This assumption is supported by MSCI’s analysis. The report observes, inter alia, that close economic integration within regions means that markets within each geographical region are more closely correlated with each other than with markets in other regions. Emerging markets however differ from developed markets in that returns are more volatile, whilst risk is more affected by country-specific factors.

A challenge in determining the adjustment factors is that equity market risk and return are multidimensional and affected by a number of factors beyond regional affiliation. MSCI’s analysis shows that risk and return in global portfolios are affected by developments in countries and industries, as well as development in several other risk factors. Another challenge is that for companies with global operations and value chains, the risk is likely to be relatively less affected by developments in the country or region in which the company is listed. The Ministry also recognises that correlations between regions fluctuate over time, and that future developments are uncertain. These factors make it challenging to assess how risk and return in a global portfolio are affected by different regional distributions of the investments.

The Ministry has noted that MSCI’s analysis shows that the benefits of a regional distribution deviating from a free float-adjusted index in the short run have been moderate. Hence, many considerations would appear to be accommodated by such an index. MSCI however emphasises that a free float-adjusted index entails a higher country concentration than that of other weighting principles. The risk associated with country concentration is not adequately captured by short-term risk measures. According to MSCI, concentration increases the vulnerability of investments to developments in individual regions and reduces the diversification of portfolio exposure to regional differences in long-term company performance. The Ministry agrees with the position that long-term investors should consider such implications.

The Ministry would however highlight that MSCI’s analysis is based on a time period (1994–2019) of intense globalisation during which country-specific considerations may have been less prominent. The extent to which the analysis of this period adequately captures future developments is uncertain. Developments towards increased economic protectionism and geopolitical uncertainty may imply larger impacts of country-specific consideration on return differences in coming years.

The coronavirus outbreak is another example of an event not captured in the time period analysed by MSCI. The chapter 9 discussion article summarises the findings from multiple academic studies of the relationship between pandemics, economic growth and financial markets. Previous pandemics have been characterised by large differences in mortality between countries and differences in economic implications both at country and sector level, in the short and long run. Pandemics in the past have caused long-term decline in the return on capital, which is a measure of profitability, as a reduced supply of labour has increased wage growth and thus reduced company profits. Studies of the initial financial market impact of the coronavirus outbreak show that both company characteristics and sector affiliations are relevant in explaining variations, although a number of country-specific considerations also appear significant. Equity markets in low-income countries have dropped more in value than equity markets in high-income countries. Equity markets in countries with elevated government debt and a large share of their population aged above 65 have also declined relatively more in value than equity markets in other countries. However, there remains considerable uncertainty as to the effects in the longer run. The government-coordinated shutdown of economic activity and the subsequent monetary and fiscal policy response during the current coronavirus outbreak are without historical precedent and merit caution when assessing the relevance of experiences from previous pandemics.

Concentration risk

The Ministry’s review of the regional distribution of the equity benchmark in 2012 emphasised that North America accounts for a large share of a free float-adjusted index. The Ministry acknowledged that such country concentration might make the benchmark index vulnerable to developments in the US economy.

The Ministry recognises that the high weight allocated to North America reflects a number of factors; see box 3.2 for further discussion of the regional distribution of global market values. The economies in North America are large and well-functioning, financial markets well-developed and broadly diversified, and a large portion of globally leading companies are listed in the region. Another factor is that listed companies in North America on average have been significantly more profitable than companies in other regions, and therefore have a higher market value.

The Ministry finds it challenging to assess the risk associated with country concentration and the extent to which such risk would merit benchmark index adjustments. The Ministry has taken notice of Norges Bank’s observation that the probability of extraordinarily weak developments in company earnings within individual markets is difficult to quantify, and that it is difficult to assess whether such risk is already reflected in market prices. Another issue is that the factors influencing corporate profitability and valuations are multifaceted. High profitability and valuations are not general market characteristics that apply to all companies in a region. It is therefore uncertain how accurately decisions on regional distribution can manage risk in a global equity portfolio. MSCI’s analysis shows that the high profitability in North America is primarily associated with individual sectors, and that there are large differences in profitability and valuation between companies within the region. These differences have increased over time and are larger in North America than in other regions.

The Ministry is nonetheless of the view that weight should be attached to country concentration when reviewing the geographical distribution of the equity benchmark. The Ministry recognises that the index provider’s free float-adjustment is a key reason for country concentration in such indices being significantly higher than otherwise implied by full market weights.

In the current GPFG benchmark index the low weighting of North America (relative to a free float-adjusted index) is mirrored by a higher weighting of European developed markets. At present, the benchmark index has a significantly higher share in Europe than the region’s share of global GDP, full market values and free float-adjusted market values. The elevated share in Europe is a result of foreign exchange risk historically being accorded significance in determining the regional composition of the benchmark. The Ministry acknowledges that the high European equity market share is an active geographical distribution choice which makes the benchmark index less representative of the global stock market’s sectoral and company composition. This choice implies, for example, that the technology sector is underweighted in the GPFG benchmark index, and that ownership stakes in European technology companies are more than twice the size of those in North American companies. The benefits of a lower country concentration in North America must be considered against such effects.

Other issues

The combination of a long investment horizon and limited liquidity requirements may, if considered in isolation, make the Fund well placed to carry certain risks and harvest potential risk premiums, for example risk premiums relating to low liquidity. Such risk premiums may give rise to differences in expected return between regions. MSCI states that the expected equity premium in emerging markets is higher than in developed markets, compensating investors for higher risk in these markets. In its letter of 21 August 2019, Norges Bank states that higher market risk in emerging markets may give reason to expect higher returns in such markets. The Bank however acknowledges that actual returns have been lower than could reasonably be expected, in view of this higher market risk. The Ministry has taken notice of Norges Bank’s recommendation to continue the current emerging market weighting in the index.

Differences in investor conditions between various markets, such as tax and corporate governance practices, may also affect expected risk and return. In its letter, the Bank points out that such factors may be of relevance to the geographical distribution if the Fund has significant advantages or disadvantages relative to the marginal investor. The Bank however recognises that it is challenging to assess this in practice, and does not recommend any specific modifications.

Proposed new adjustment factors

Based on an overall assessment, the Ministry is proposing to modify the adjustment factors for the GPFG equity benchmark, somewhat reducing the overweight in developed markets in Europe and correspondingly increasing the share invested in North America. This is in line with the recommendations of Norges Bank. The purpose of this modification is to ensure that the equity benchmark better reflects economic activity in listed companies globally. However, the requirement for broad diversification across various regions suggests that the North American share of the benchmark index should be somewhat lower than implied by free float-adjusted market weights.

The Ministry’s proposed new adjustment factors are set out in table 3.3. It is estimated that the weight to developed markets in North America will increase by about 6 percentage points with new adjustment factors, to just over 48 percent, based on market values at year-end 2019. The weight to developed markets in Europe is reduced more or less correspondingly, but will still remain higher (by about 8 percentage points) than the region’s share of the global listed stock market (full market weights) and the region’s weight in the free float-adjusted FTSE Global All Cap index.

No change is recommended to the adjustment factor for other developed markets. The index weight to other developed markets is not materially different from this region’s share of the global stock market (full market weights) or free float-adjusted market weights. The Ministry is further proposing to maintain the current adjustment factor for emerging markets. The continuation of the adjustment factors for other developed markets and emerging markets is in line with Norges Bank’s recommendation.

The Ministry’s proposed new adjustment factors will make the benchmark index more representative of the global stock market and somewhat less closely tied to development in European companies. The Ministry expects that the return on the benchmark index will be more in line with that implied by global equity market developments, represented by free float-adjusted indices. The new adjustment factors are not anticipated to have significant implications on expected risk and return for the Fund as a whole. The properties of the new equity benchmark, such as historical risk and return, sectoral distribution and ownership stakes, are discussed in box 3.3.

The operational implementation of the new regional distribution of the equity benchmark is projected to involve significant transaction volumes. A gradual phase-in will contribute to lower transaction costs. The coronavirus pandemic means there is currently higher financial market uncertainty than during normal times. The phase-in plan should reflect this uncertainty. The Ministry will prepare a plan for phasing in the new equity benchmark in consultation with Norges Bank.

The emerging market sub-index

The GPFG equity benchmark, as a general rule, includes all markets classified by the index provider FTSE Russell as developed, advanced emerging or secondary emerging markets. The index provider has constructed the index on the basis of transparent, rule-based and verifiable criteria.

The benchmark index has over time been expanded to include additional emerging equity markets, from a small number in 2000 to 23 at year-end 2019. In addition to the index provider’s market classification, the Ministry requires the Executive Board at Norges Bank to approve all markets in which the equity portfolio is to be invested. The Bank has to date approved all markets in the benchmark index, with the exception of Pakistan. The Bank has further chosen to invest the Fund in markets and companies outside the benchmark. Such investments are made within the scope for deviations from the benchmark index.

Emerging markets constitute a significant part of the global equity market, cf. figure 3.2. Investments in these markets may contribute to diversification and harvesting of risk premiums. However, emerging markets differ from developed markets in returns being more volatile and risk more country-specific. Emerging markets are less homogeneous than developed markets and are generally characterised by weaker institutions, less transparency and weaker protection of minority shareholders. Such matters make adequately maintaining the role as a responsible investor more challenging. In its letter of 22 August 2019, Norges Bank however states that its experience from nearly 20 years as a state-owned minority shareholder in emerging equity markets is positive.

In its recommendations on the Government Pension Fund white paper in the spring of 2018, the Standing Committee on Finance and Economic Affairs commented that when the benchmark index is expanded to include countries in which the human rights situation gives cause for concern, it is important for the guidelines for observation and exclusion to be reviewed and «operationalised in a manner that ensures a sufficiently robust framework to address the challenges in these markets»9. Furthermore, the Council on Ethics in a letter of 13 November 2018 stated that several countries included in the benchmark index pose a high risk of human rights violations, and that the access to information in these countries is generally limited. This makes responsible investment more challenging.10

The Ministry of Finance deems these considerations to be of relevance to the assessment of emerging market investments. On this point, the Ministry in 2019 appointed a committee to review the ethical guidelines for the GPFG. The committee has, among other things, examined issues relating to individual countries, including countries whose local regulations violate recognised international conventions and standards, as well as countries where access to information is limited. The committee submitted its report on 15 June 2020. The committee’s conclusions will form part of the assessment basis for the Ministry’s review of the emerging market sub-benchmark. The report has been circulated for public consultation, and the Ministry intends to present its analysis and recommendation in the white paper on the Government Pension Fund in the spring of 2021. The committee’s report is discussed in chapter 6.3 (in Norwegian only).

The Ministry finds there is need for further review of the emerging market sub-index, including to what extent emphasis should be placed on the distinctive characteristics of such markets when determining the composition of the benchmark index. Norges Bank has in its letter of 22 August 2019 stated that the current method and rules for the composition of the emerging market equity sub-index should be maintained. The significance of country-specific considerations in emerging markets suggests that one might consider putting a cap on the share of the benchmark index for which any one emerging market may account. Special emphasis may be put on such considerations if high exposure is accompanied by high risk, including non-financial risk. In a letter of 15 April 2020, the Ministry asked the Bank to consider the specifics as to how such a limit might be structured, as well as the associated implications. The Ministry received the Bank’s analysis and assessments on 27 August 2020. The letter is available on the Ministry website.

Against this background, the Ministry intends to revert with its recommendation on the emerging market investment framework in the white paper on the Government Pension Fund to be presented in the spring of 2021. This will enable the Ministry to draw on the Ethics Committee’s report and consultative comments thereon, in addition to analysis and advice from Norges Bank.

The Ministry will not include any new markets in the equity benchmark until the framework for and composition of emerging market investments have been concluded. This means that Saudi Arabia, which FTSE Russell included in FTSE Global All Cap in March 2019, and Romania, which is scheduled for inclusion in September 2020, will not be included in the Fund benchmark index until further notice.

Textbox 3.2 The regional distribution of global market values

The regional distribution of global free float-adjusted market values deviates from the distribution implied by the relative size of the economies as measured by gross domestic product (GDP). The difference is most prominent in North America and in emerging markets. Whilst the weight attributed to North America in a free float-adjusted index is significantly higher than would be implied by the region’s share of global GDP, the opposite applies to emerging markets. Figure 3.2 provides an overview of the regional distribution of global GDP, as well as the distribution of companies’ book equity, profits, free float-adjusted and total market values. The analysis is based on the MSCI ACWI equity index. An analysis based on index products from the provider of the GPFG equity benchmark, FTSE Russell, would have produced broadly similar findings.

Zoom in on image

Source MSCI.

North America

The large North American share of the global stock market measured by market capitalisation is driven by companies in this region on average having been more profitable than companies in other regions. The region’s share of the global equity market is somewhat larger than its share of global GDP, whilst high return on equity has caused North America’s share of global profits to be considerably larger. Return on equity is an important reason why valuations for North American companies are higher than valuations in other regions. The region’s share of the global equity market is therefore larger than would be suggested by its share of both profits and book equity. Free float-adjustment serves to further increase the high weighting of North American equities, from 47 percent at full market weights to 58 percent in the MSCI All Cap index.

Box 3.2 (cont.)

There may be several reasons why the return on equity of North American companies on average is higher than that of companies in other regions. It has, among other things, been suggested that large companies in the region may hold considerable market power.¹ Developments towards higher market concentration in several goods and service markets have been highlighted in this context. Moreover, companies such as Google and Facebook have established a dominant position in their respective markets. Higher profitability may at the same time be the result of cost advantages. MSCI’s analysis of global equity returns shows that the situation is complex. There are also large return variations within the region, and the high return on equity can be attributed to the technology, manufacturing and consumer goods sectors.

Emerging markets

Companies listed on regulated marketplaces account for a smaller portion of economic activity in emerging economies than in developed markets. The book value of listed companies domiciled in emerging markets therefore represents a smaller share of global equity markets than emerging markets’ share of the world economy as measured by GDP; see figure 3.2. The equity return of companies listed in emerging markets is on a par with companies in developed markets, but their valuations are nonetheless lower. Higher risk is an important reason why emerging market company valuations are lower than valuations of developed market companies. The analysis from MSCI shows that emerging markets are characterised by higher market risk, lower liquidity and higher country risk.

The free float-adjustment made by the index provider has substantial impact on emerging markets’ share of the index. The emerging market share is reduced from 23 percent at full market values to 12 percent following the adjustments. A high proportion of companies controlled by states, families or other long-term owners explain the large free float-adjustments in emerging markets. Concentrated ownership may have implications for corporate governance and minority shareholder rights. MSCI finds that sizable free float-adjustments typically coincide with weaker average corporate governance.

¹ See for example Chad Syversen (2019), «Macroeconomics and Market Power: Context, Implications, and Open Questions», Journal of Economic Perspective, 33 (3) pp. 23–43.

Textbox 3.3 Properties of the new equity benchmark

Historical risk and return

The GPFG equity benchmark is based on the global index FTSE Global All Cap. The Ministry has grouped the equity markets of the benchmark index into four regions: developed markets in North America, developed markets in Europe, other developed markets and emerging markets. Norway is excluded from the benchmark. The Ministry has further assigned adjustment factors based on regional affiliation, modifying the free float-adjusted market values of the companies in the four regions. The adjustment factors result in European companies having a higher weight in the equity benchmark than in the FTSE Global All Cap index, whilst North American companies are given a lower weighting. Companies in Asia-Pacific (developed markets) and in emerging markets have more or less the same weighting as in FTSE Global All Cap.

The adjustment factors have resulted in the actual risk (volatility) and return of the equity benchmark deviating from the risk and return of the FTSE Global All Cap index (excluding Norway). This is primarily caused by different weighting of developed markets in North America and Europe.¹

Table 3.4 presents annualised return and risk measurements for the FTSE Global All Cap index over the period September 2003 to December 2019, measured in the GPFG currency basket. The table also includes values for an adjusted FTSE Global All Cap index, in which the current and new adjustment factors, respectively, have been applied to the four regions, henceforth referred to as the current benchmark index and the new benchmark index.

Table 3.4 Historical annual returns, volatility and maximum drawdown for the FTSE Global All Cap index, the current benchmark index and the new benchmark index over the period September 2003 to December 2019, measured in the currency basket of the Fund. Percent

	FTSE Global All Cap	FTSE Global All Cap with current adjustment factors	FTSE Global All Cap with new adjustment factors
Return	9.11	8.77	8.90
Volatility	13.0	13.4	13.2
Maximum drawdown	-55.0	-56.4	-55.8

Source Thomson Reuters Datastream and the Ministry of Finance.

The table shows that the FTSE Global All Cap index has delivered somewhat higher return and lower volatility than the other indices. This is primarily the result of a higher weighting to North America, which over the period analysed was characterised by a higher average return and lower volatility than European markets. The new benchmark index would have generated a somewhat higher return and lower volatility than the current benchmark index.

However, the return difference has varied over time. This may be illustrated by calculating the deviations in rolling periodical returns. Figure 3.3 shows the difference between annualised 5-year return on the new benchmark and annualised 5-year return on the current benchmark. As the figure illustrates, the current benchmark would have delivered a higher return than the new benchmark until year-end 2010. This finding implies that it would have been preferable to hold a higher share in Europe up until and including 2010. The opposite has been the case since 2011 and onwards. It can reasonably be assumed that return differences between the current and new benchmark indices will continue to vary over time.

Maximum drawdown denotes the maximum reduction from peak to trough over a chosen period, and is thus a risk measure for large reductions in value («downside risk» or «tail risk»). Table 3.4 illustrates that this risk has been at about the same level for the current and new benchmark indices. The maximum drawdown measured in the period was -56.4 percent and -55.8 percent, respectively. This drawdown was registered over the period 31 October 2007 to 27 February 2009.

Source Thomson Reuters Datastream and the Ministry of Finance.

Tracking error

The annual standard deviation of differential return is referred to as «tracking error». Tracking error is a measure of differential return volatility over time. The tracking error of the new benchmark index measured against the current benchmark index is calculated at 0.6 for the period September 2003 to December 2019, measured in the currency basket of the Fund. The statistical interpretation of this finding is that the annual return difference has been less than 0.6 percentage points in about two of three years over this time period.

Sectoral composition

The new adjustment factors imply that developed markets in North America will constitute a larger share of the benchmark index mirrored by a smaller share in European markets. This further entails minor changes in the index sectoral composition; see figure 3.4. The technology sector’s share of the benchmark index will increase by about one percentage point. The change in other sectors is less than 0.3 percentage points. The new sectoral composition is shown in figure 3.5. The technology sector will account for about 15 percent of the new benchmark index.

Zoom in on image

Source Thomson Reuters Datastream and the Ministry of Finance.

Zoom in on image

Source FTSE Russell and the Ministry of Finance.

Ownership stakes in companies

The new adjustment factors imply that average ownership stakes will be somewhat more equal across regions. Table 3.5 presents average, free float-adjusted ownership stakes at year-end 2019 under the current and new benchmark index, respectively. Ownership stakes in North American companies will increase by about 0.15 percentage points, whilst ownership stakes in European companies will be reduced by close to 0.5 percentage points. Changes for companies in Asia-Pacific and emerging markets will be marginal.

Table 3.5 Average, free float-adjusted ownership stakes with current and new adjustment factors. Values as per 31 December 2019. Percent

	Europe (developed)	North America (developed)	Other (developed)	Emerging markets
Current ownership stake	2.50	1.01	1.58	1.48
New ownership stake	2.01	1.16	1.58	1.49

Source Thomson Reuters Datastream and the Ministry of Finance.

¹ Minor differences in the weighting of the regions Asia-Pacific (developed markets) and emerging markets, as well as ethical exclusions, have also contributed to risk and return deviations, but the effect is limited.

3.2 Investments in companies seeking a listing

3.2.1 Background

In the Government Pension Fund 2018 white paper, the Ministry of Finance discussed whether to allow for the GPFG to on a general basis be invested in unlisted equities. The Ministry acknowledged that allowing for unlisted equity investments may provide Norges Bank with additional investment opportunities in its active management. At the same time, this type of investment may affect the Fund’s reputation and challenge key characteristics of the current investment management framework, such as transparency, low costs and an emphasis on systematic risk in listed markets. It was observed that unlisted equity investments generally are more complex, requiring different expertise than listed equity investments, and demanding a different form of governance structure and lines of responsibility on part of the investment management organisation. The Ministry thus proposed not to allow for the inclusion of unlisted equity investments in the GPFG. This was endorsed by the Storting; see Innst. 370 S (2017–2018)(in Norwegian only).

In the white paper, the Ministry of Finance made reference to Norges Bank’s existing permission to invest the GPFG in unlisted companies whose board of directors has expressed an intention to seek a listing, and stated that this permission would be further discussed in the dialogue with the Bank. On this matter, in a letter of 27 March 2019, the Bank was asked to, among other things, provide an account of its framework for, and experience with, investing in companies seeking a listing, as well as to assess whether the regulation in the mandate stipulated by the Ministry should be amended. The Ministry received the Bank’s reply letter on 23 August 2019; see section 3.2.2. The letters are available on the Ministry website. The Ministry’s assessments are outlined in section 3.2.3.

3.2.2 Norges Bank’s letter

Norges Bank refers to that, as for other investments, the Bank’s framework for investing in unlisted companies seeking a listing requires procedures to be in place for the management, control and monitoring of relevant risks. Among other things, the Bank emphasises that the investment instrument, and the market in which the company is planning to list, must be approved before the investment is made, cf. Section 3-10 of the mandate. The Bank also notes that a thorough due diligence review, addressing any financial, legal, tax and commercial matters of significance, is conducted prior to making any investment.

Norges Bank states that although it has considered several opportunities, it has completed only one investment in a company seeking a listing. In May 2012, the Bank entered into an agreement for the purchase of 4.2 percent of the company Delta Topco Ltd., which held the Formula 1 rights. At the time of purchase, the company board had applied for a listing in Singapore. In June 2012, the board decided to postpone the listing due to changed market conditions. The Bank chose to retain its investment in the company until 2017, when the position was sold to Liberty Media in return for listed shares of said company.11 According to Norges Bank, Delta Topco Ltd. paid large dividends to the Bank during its period in private ownership, and that overall return on the investment was in line with the general equity market when differences in market risk are taken into account.

Norges Bank states that the handling of this investment illustrates how the Bank will normally manage a situation in which the company does not go ahead with the planned public offering. The option to retain the ownership stake in the unlisted company is highlighted as preferable to a situation in which the Bank is forced to divest the share within a specific time frame. The Bank is of the view that the established framework for such investments has worked well.

The Bank further states that the Fund often participates in initial public offerings (IPOs), as well as transactions following an IPO. Participation in IPOs may contribute to more efficient investment management by investing the Fund in shares prior to benchmark inclusion of such shares. However, not all shares acquired in IPOs will subsequently be included in the benchmark index. The Bank notes that more than half of IPOs where the Fund has participated since 2016 have not been followed by benchmark inclusion of the shares in question. Participation in IPOs has over time generated excess return relative to purchasing the same shares immediately after listing.

Norges Bank has thus far interpreted the provision on intention to seek listing to mean that the company needs a board resolution to seek listing in the near future. In the Bank’s experience, such resolutions are typically adopted at a late stage in the process, and a number of transactions that might have been of interest to the Fund have therefore not been considered. At the same time, the experience from the Delta Topco Ltd. investment demonstrates that a listing may not necessarily materialise despite of the company board having filed a formal listing application. The Bank is of the view that the wording of the current provision, which requires the board of directors to have expressed an intention to seek a listing, should be amended.

According to the Bank, an alternative approach might be to allow for the GPFG to be invested in shares of large companies that are not yet listed, with such investments being limited to a maximum of 1 percent of the equity portfolio. Such companies often have other institutional investors among the owners, which may generate some liquidity in the share. The Bank finds it reasonable to expect a number of these companies to seek a listing at a later date. The probability of listing is one of the factors the Bank would have to assess during the due diligence reviews. The suggested amendment of the mandate provision would in the view of the Bank enable it to build, over an extended period of time, the desired exposure in large companies that have yet to be listed.

The Bank comments that such investments should be regulated in the same manner as other investments not included in the benchmark index. The regulation should be structured such as to enable the Bank to manage the investments in a way that supports the investment management objective. The Bank emphasises it should be permitted to hold the unlisted companies for as long as deemed appropriate by the Bank.

Norges Bank states that clearer regulation of investments in large companies that have yet to be listed may serve to adequately control the risk associated with such investments. The regulation should, according to the Bank, involve a combination of limits stipulated by the Ministry of Finance and the Executive Board of Norges Bank, respectively. The Bank believes that it among other things should be delegated to the Executive Board to stipulate the minimum size of companies eligible for investment, as well as caps on ownership stakes in individual companies.

3.2.3 The Ministry’s assessments

The management mandate for the GPFG allows Norges Bank to invest the equity portfolio in unlisted companies where the board of directors has expressed an intention to seek listing on a regulated and recognised marketplace. This access has been in place since 1 January 2011. In the white paper on the Government Pension Fund in the spring of 2011, The Ministry of Finance commented that this will enable the Bank to take advantage of the GPFG distinctive characteristics, whilst ensuring efficient and effective execution of the investment management mission.

The intention with the mandate provision is to enable the Bank to build exposure to companies in advance of an upcoming initial public offering. The Ministry emphasises that there generally is considerable uncertainty as to whether and when companies will be listed. The Ministry finds it reasonable to assume that there is less uncertainty about whether a listing will take place with a board resolution on listing than without such a resolution. The Bank’s proposal would, in the view of the Ministry, entail a higher probability that the GPFG will remain directly invested in unlisted companies for a long time without any reference to a specific board resolution as the rationale behind the investment decision.

The Ministry is of the view that the current requirement for an intention to seek a listing should be maintained. The mandate regulation of such investments should, however, be clarified somewhat.

Unlisted investments are generally more complex than listed investments and not subject to the same reporting requirements. Such investments are not traded in the market on a regular basis, and it is more challenging to establish the actual market price of unlisted investments. Risk and return assessments will thus be more uncertain than in listed markets. Low liquidity implies that disposing unlisted holdings may be more difficult and could impose losses on the Fund, especially if divestment needs to be completed within a limited time period.

The mandate regulation of investments in unlisted companies seeking a listing should therefore, in the view of the Ministry, be modified into closer alignment with the regulation of other unlisted investments. Norges Bank emphasises that, in accordance with the Bank’s investment framework, thorough due diligence reviews are conducted prior to any investment. However, this is not a requirement under the mandate stipulated by the Ministry. The Ministry deems it appropriate to formalise the due diligence requirement for investments in unlisted companies seeking a listing. It is thus proposed to amend Section 3-10, Sub-section 4, of the mandate on thorough due diligence reviews prior to any investment in the unlisted real estate portfolio and renewable energy infrastructure portfolio to also encompass unlisted companies seeking listing. Comprehensive due diligence reviews in advance of each investment in unlisted assets shall seek to identify all relevant investment risk, as well as operational and non-financial risk, including potential reputational implications of the investments. The mandate Section 2-5, Sub-section 6, on the Executive Board’s approval of individual investments in the unlisted portfolios in excess of certain threshold amounts, is further proposed extended to also encompass investments in unlisted companies seeking a listing.

The Ministry concurs with Norges Bank’s assessment that it may be beneficial to allow for the Bank to retain unlisted holdings in the event that the listing does not go ahead, rather than the Bank being forced to divest within a limited time frame. A specification in the mandate that the Executive Board shall issue guidelines on the disposal of shares in unlisted companies whose planned listing does not take place will in the Ministry’s view establish that the Fund may remain invested in the relevant company for as long as deemed appropriate by the Bank.

3.3 Climate risk

3.3.1 Introduction

Climate change will affect companies and future economic development. International collaboration will be of key importance in tackling climate change. Many countries, including Norway, have committed to considerable greenhouse gas emission reductions under the United Nations Framework Convention on Climate Change and the Paris Agreement. A reduction in total global emissions in line with the Paris Agreement will have major impact on companies and future economic development, and will require comprehensive measures in several areas, including climate policy, technology and investment.

It is however challenging to analyse climate change implications for individual companies, industries and markets, and thereby for investors. There is considerable uncertainty as to the physical effects of climate change and associated financial risks, as well as the economic implications. In addition, there is considerable uncertainty regarding climate policy and technological developments in the transition to a low-carbon economy. This uncertainty entails significant risk, which needs to be managed by investors.

In well-functioning financial markets, capital will be allocated to investment projects and companies generating the highest expected risk-adjusted return. However, lack of available and relevant information on the climate risk facing companies indicates that it may be difficult for financial markets to price this type of risk. This may result in financial markets not allocating capital to businesses best placed to manage climate risk, or to companies or projects with superior prospects for sustainable economic activity.

In order for financial markets to efficiently facilitate the transition to an economy with lower greenhouse gas emissions, market participants require knowledge and information about the risks as well as the opportunities arising from climate change, climate policy and new technology. In recent years, several initiatives have therefore been launched to promote more well-informed investment decisions. Such initiatives include frameworks intended to ensure more standardised and comparable reporting on the risks climate change and climate policy changes pose to companies and business models.

Financial market participants are increasingly focusing on responsible investment practises and sustainability, and seek knowledge about the implications for their activities. Investors are concerned with how sustainability and climate risk considerations may be integrated into investment management decisions. Knowledge of the risk posed by climate change, for both individual investments and specific geographical areas, is continually improving, but financial market participants are still facing general challenges regarding access to relevant, high-quality data that can be systematically applied.

The Ministry of Finance and Norges Bank have, like other market participants, placed emphasis on how climate risk may affect financial assets. Climate risk is an integrated part of the Bank’s investment management. Norges Bank’s annual publications on responsible investment provide a thorough presentation of the climate risk efforts of the Bank, as part of the wider work on responsible investment. These efforts have been discussed in several white papers, including the white paper on the Government Pension Fund published in the spring of 2017. In the Government Pension Fund 2018 white paper, the climate risk reporting framework was specifically addressed, referring in particular to the recommendations of the Task Force on Climate-related Financial Disclosure (TCFD)12. The Ministry expressed an expectation that Norges Bank will continue its climate risk reporting efforts in line with the recommendations of the task force, among other things by contributing to the development of standards and reporting practises.

In the white paper on energy equities in the GPFG, the Ministry commented that climate risk is an important financial risk factor for the GPFG. It was also referred to the group of experts which examined energy equities in the Fund and recommended the Ministry to consider reviewing the climate risk efforts in order to assess whether further strengthening is warranted. Against this background, The Ministry of Finance proposed requesting Norges Bank to review and describe said efforts. The Storting endorsed the Ministry’s proposal.

Norges Bank was in a letter of 28 June 2019 asked to describe internal processes and methods for assessing climate risk, including the various tools used in this regard. The Bank was further asked to explain how climate risk in the equity, fixed-income and real estate portfolios is assessed, as well as how individual companies with the highest contributions to overall climate risk in the Fund are assessed. Furthermore, the Bank was asked to describe how climate risk is integrated into investment decisions, as well as any plans for further progressing its climate risk efforts within the limits of the established investment strategy for the Fund and the mandate stipulated by the Ministry of Finance.

Norges Bank has in a letter of 26 November 2019 to the Ministry provided an account of its efforts on climate risk; see the description in section 3.3.2.13 The discussion of the efforts of Norges Bank is also based on the Bank’s responsible investment report for 2019. The Ministry’s assessments are outlined in section 3.3.3.

3.3.2 Norges Bank’s efforts

In its letter, Norges Bank emphasises that the Bank manages climate risk within the mandate limitations set by the Ministry, and that adapting or applying the Fund portfolio for the achievement of any specific climate policy targets does not form part of the mandate.

Norges Bank comments that climate risk is one of several financial risk factors for the GPFG. Climate risk is considered a complex risk factor, and it is anticipated that the understanding of the financial implications for the Fund will evolve over time. The prices of the assets in which the fund capital is invested, and the extent to which these prices reflect climate risk, will affect the financial risk of the Fund. Norges Bank comments that a broadly diversified and market-weighted portfolio like the GPFG will be exposed more or less to the same financial climate risk as the underlying markets and industries in which it is invested.

Analysis methods

Norges Bank uses scenario analysis to improve its understanding of how climate risk may affect individual companies and the Fund portfolio. Such analysis may shed light on both physical risk14 and transition risk15, although this analysis will be subject to considerable uncertainty. The scenario analysis may serve to illustrate potential outcomes, but does not represent predictions for the future.

The Bank has developed internal scenario analysis models of the equity portfolio, which are used to analyse future cash flows and CO₂ emissions at the company level, including how potential future regulations, such as carbon pricing and carbon emission allowances, may affect various companies, industries and regions under different climate scenarios towards the year 2100. Provisional analysis indicates that certain companies in sectors like manufacturing, energy, raw materials, utilities and consumer goods may face financial challenges as the result of future carbon pricing.

In 2018, the Bank joined a pilot project along with other global investors under the United Nations Environment Programme (the UNEP Finance Initiative), the purpose of which is to develop investor reporting methods based on the TCFD framework. Experience from this project has demonstrated that additional effort is required to enable companies and investors to fully implement the TCFD recommendations. The summary report from the project includes an overview of various scenario analysis methods and tools, and a review of leading analysis providers. The project included an analysis of potential climate outcomes’ effect on the equity portfolio of the Fund. The portfolio «Value at Risk» was estimated by examining the implications regulatory risk and technological developments associated with a 2-degree scenario and extreme weather conditions may have for corporate’ revenues.

Measurement of corporate greenhouse gas emissions

Norges Bank estimates the amount of greenhouse gas emissions from the Fund investee companies, based on emissions data reported directly by the companies and third-party estimates in cases where the companies do not provide adequate reporting. The Bank comments that limited access to relevant, high-quality data is a general challenge.

The Fund’s total carbon footprint, measured as the total greenhouse gas emissions from the portfolio companies, may shed light on risk and opportunities across industries. It may also provide a basis for assessing how changes in the portfolio may affect the overall carbon footprint. However, carbon footprint calculations provide only a snapshot and do not consider industrial structure, company strategy and other factors. Nor do carbon footprint calculations include (indirect) emissions through the production chain16.

The Bank’s carbon footprint analyses encompass both equities and corporate bonds, and are based on the TCFD framework recommendations for asset managers. Calculations based on data for 2018 show that greenhouse gas emissions in the equity portfolio are largely driven by companies in fossil fuel power generation and energy-intensive industries, such as raw materials, metals production, and heavy industry.17 Within these high-emission sectors, the largest companies account for the largest emissions. Company exclusions based on the ethical guidelines (primarily the product-based coal criterion) have according to the Bank’s calculations served to make estimated greenhouse gas emissions for the Fund portfolio 14 percent18 lower than would have been the case without such exclusions.

Norges Bank further estimates the emission intensity of the equity and corporate bond investments of the Fund. This is calculated as the number of tonnes of CO₂ equivalents per million US dollars in revenues, and allows comparisons across companies and industries as CO₂ emission levels are scaled by revenues. Estimations show that the emission intensity of both the equity portfolio and the corporate bond portfolio is lower than that of their respective benchmark indices. However, these differences are not the result of specific investment strategies, but rather reflect a number of individual investment decisions.

Norges Bank also recognises that the Fund’s unlisted real estate investments are directly exposed to climate risk, in the form of both physical risk and transition risk. The exposure to such risks is primarily affected by the location of the properties and whether their construction and operational facilities make them well suited to withstand impacts of climate change and expected future regulatory measures. Properties in flood-prone areas are often subject to higher insurance premiums and stricter regulatory flood-proofing requirements. The Bank states that local authorities in the eight cities in which the Fund holds unlisted real estate investments have developed long-term greenhouse gas emission reduction plans that will affect the real estate sector. This exposes the Fund’s real estate investments to regulatory risk. The Fund is also exposed to market risk associated with declining demand for buildings without green qualities and energy-efficient operation.

Integrating climate risk into investment decisions

Norges Bank states that climate risk assessments are integrated into the investment decisions on equities, bonds and unlisted real estate. Climate regulation developments and new technology are key factors in assessing the future earnings of companies in several sectors, applicable to both the equity and bond portfolio. The Bank has developed a framework for systematic assessment of investment risk and operational risk for various sovereign issuers, which includes assessments of environmental factors such as emission intensity and climate change exposure. For unlisted real estate, thorough due diligence and analysis are carried out prior to making any investment. Such analysis includes, among other things, assessments of the extent to which the property exposes the Fund to physical risk and transition risk.

The Bank has accumulated internal environmental technology expertise through its management of the environmental investment mandates. The environmental equity investments are concentrated in companies involved in low-emission energy and alternative fuels, clean energy and energy-efficiency technology, as well as technology and services for the management of natural resources.

Sustainability assessments may result in the Fund divesting from companies viewed by the Bank to pose especially high risk, including greenhouse gas emission-related risk. The Bank recognises that companies whose operations or value chains involve excessive greenhouse gas emissions may be exposed to considerable transition risk. The Bank has during the period 2012–2018 completed risk-based divestments in 142 companies on the basis of climate risk assessments.19 These are predominantly relatively small companies that have particularly high greenhouse gas emissions or are contributing to deforestation.

The ethically motivated guidelines for observation and exclusion of companies include criteria targeting companies that contribute to, or are themselves responsible for, unacceptable greenhouse gas emissions, as well as mining companies and power producers that base their operations on coal. Companies falling within the scope of the coal criterion may be excluded or placed under observation by Norges Bank without any recommendation from the Council on Ethics. Exclusion decisions based on the coal criterion represent, according to Norges Bank, the largest individual contribution to reducing greenhouse gas emissions in the portfolio, compared to the underlying index products. See chapter 6.2 for further discussion of the ethically motivated guidelines (in Norwegian only).

Climate risk management

Norges Bank comments that active ownership is a key element of climate risk management. The Bank focuses this effort on companies in the most exposed sectors. The Bank reports to have strengthened the parts of the organisation responsible for active ownership and risk assessments relating to environmental, social and corporate governance issues.

The measures applied by the Bank in addressing climate risk are predominantly an extension of the responsible investment measures discussed in chapter 6.2. Climate risk efforts through the Bank’s active ownership are discussed in box 6.3 (in Norwegian only).

In addition to the regular measures used by the Bank in its responsible investment management, the Bank is collaborating with investment partners and property managers to integrate environmental measures in the business plans for the unlisted properties. Norges Bank has, among other things, published a guidance document on environmentally sustainable real estate management. Moreover, the Bank is developing a platform to gather data on the properties in the portfolio, including information on energy and water consumption, waste management and environmental certification. This information will be used to measure the environmental impact of the properties, to assess environmental measures and to estimate climate accounts for the unlisted real estate investments.

Norges Bank recognises that environmental certification will become increasingly important for property competitiveness. The Bank is aiming to have all office and retail properties in the portfolio environmentally certified.

Plans to further advance the work on climate risk

In its letter, Norges Bank states it will continue to support initiatives aiming to develop global reporting standards. The Bank recognises that information on greenhouse gas emissions and how company operations may be affected by climate change is of decisive importance in analysing portfolio climate risk. Access to relevant, high-quality data is a key priority. The Bank observes that numerous initiatives are seeking to increase the scope of companies’ reporting and investors’ access to data. Such initiatives include TCFD, the Global Reporting Initiative (GRI), the Sustainability Accounting Standards Board (SASB) and CDP (formerly the Carbon Disclosure Project).

The Bank will further continue its development of analytical tools for climate scenarios, and will incorporate more detailed, company-specific data such as financial data, emission figures, specific emission reduction targets and exposure to regional carbon mechanisms, in order to expand and improve the internal models for analysing different climate trajectories. In addition, the Bank is exploring methods for analysing physical climate risk in the equity portfolio.

Active ownership is a key element in managing climate risk. Norges Bank will continue the dialogue with companies in exposed sectors, and will seek to clearly communicate its expectations that investee companies manage climate risk and reduce greenhouse gas emissions.

Norges Bank comments that the work on assessing physical risk in the real estate portfolio will remain a key priority. The Bank is planning to expand the flood-risk analysis based on historical data to also include forward-looking scenario analysis. The Bank will continue to support a collaboration project on developing methods for analysing the expected decarbonisation of the property market.

3.3.3 The Ministry’s assessments

The investment objective for the GPFG is to achieve the highest possible return, given an acceptable level of risk. The Ministry of Finance stipulates general investment management guidelines, and Norges Bank manages various types of risk, including climate risk, within these limits. The management mandate includes no requirement for the investments in the Fund to be adapted to, or used for the achievement of, any specific climate target. The GPFG has a financial objective, and there is broad political consensus that the Fund shall not be a climate policy tool. Norges Bank’s climate risk management is therefore financially motivated.

Climate risk is a complex and multifaceted financial risk factor comprising both physical risk and risk associated with transitioning to a low-carbon economy; so-called transition risk. Physical risk is the risk of the Fund’s investments being affected by physical implications of climate change, such as extreme weather, flooding, draught or heatwaves. Transition risk concerns the impact on Fund investments of climate policy and technological developments upon transition to a low-carbon economy. Both physical risk and transition risk are relevant risk factors for the GPFG, but their significance will depend on which climate scenario materialises.

As a large, long-term investor with ownership stakes in several thousand companies worldwide, the Fund’s long-term return will to a great extent be determined by returns in global equity and fixed-income markets. Climate change affects global growth and company earnings, and climate risk is therefore a relevant risk factor for the Fund. This is also reflected in the management mandate for the Fund, which states that good long-term return is considered to depend on sustainable development and well-functioning markets.

There is, at the same time, considerable uncertainty as to when, how and to what extent physical risk and transition risk associated with climate change will affect economic growth and company earnings, and thus the investments of the Fund. Companies are impacted in different ways, and returns in some asset classes, sectors and companies will be more affected than others. Current knowledge is limited, the data incomplete, and the models and analysis used are subject to considerable uncertainty. This makes it challenging to assess the specific financial implications of climate change.

It is important that Norges Bank approaches the financial risk resulting from climate change in a holistic and systematic manner. This means that the Bank should continue to integrate climate risk assessments in its risk management, investment decisions and active ownership. It is also important for this work to be based on an overall strategy, which includes prioritisations, specific objectives and planned activities to attain said objectives.

An important part of the Bank’s efforts on climate risk should, in the view of the Ministry, be to promote enhanced understanding and knowledge of climate change as a financial risk factor. Research contributions and support for initiatives aiming to develop global reporting standards on sustainability and climate risk are important for establishing common frameworks and standards, and for progressing the understanding of climate risk.

Future climate change is subject to considerable uncertainty. Such uncertainty needs to be considered from a risk perspective. Scenarios may in many cases be used to provide insights into the uncertainty associated with climate change and developments in climate policy and technology. It is positive that Norges Bank is continuing its development of analytical climate scenario tools, as well as expanding and improving internal models for analysing different climate trajectories. The Ministry would expect Norges Bank to the extent possible to stress test various climate scenarios in line with the TCFD recommendations, including a scenario in which temperature increases are contained in accordance with the ambitions of the Paris Agreement.

It is uncertain to what extent investors will be able to make gradual adaptations to future climate policy changes. A climate policy that is compatible with the Paris Agreement will require swift and significant restructuring. The uncertainty as to how financial markets can and will react to such regulatory changes adds to the importance of seeking knowledge on different climate scenarios and associated effects on financial assets in order to manage this risk in the best possible manner.

The Ministry would in this context emphasise the importance of considering potential outcomes that may have significant implications for the Fund if such a scenario materialises. These include, among other things, scenarios in which climate change crosses critical tipping points, i.e. the triggering of self-reinforcing processes and thereby major changes,20 as well as scenarios involving high-impact regulatory measures and technological breakthroughs.

The Ministry has noted that available climate scenarios and underlying analysis models are subject to considerable uncertainty. The availability and quality of data however remain challenging. It is important for climate reporting to be understood in the appropriate context and for Norges Bank to clearly communicate the limitations and uncertainties associated with the models and data.

The Ministry agrees with Norges Bank that active ownership should be a key tool in managing climate risk in the Fund. This is highlighted by the long-term and systemic nature of such risk. Norges Bank may influence company decisions and developments in a number of ways, including through voting, company dialogue and collaboration with other investors.

Climate risk efforts in the GPFG need to reflect the ambition that responsible investment practices in the GPFG shall be in line with best practice internationally. The Ministry expects the Bank to continue considering a range of active ownership measures, including participation in research projects and investor initiatives aimed at developing better models and methods, as well as standards and market practices, for the assessment and management of climate risk. Furthermore, the Ministry expects Norges Bank to continue to communicate expectations that companies plan for different climate scenarios and assess the climate risk in their operations.

Climate risk management requires knowledge and expertise on climate change and the implications of such change. Norges Bank may, within limits stipulated by the Ministry, choose a different composition of the investments than would be implied by the Fund benchmark index. To tilt the composition of an investments portfolio in order to factor in various types of risk, including climate risk; so-called active management, is challenging in practice and requires detailed market knowledge. Such decisions are therefore delegated to Norges Bank.

The broad approach of Norges Bank’s climate risk efforts may give rise to interaction effects in the form of expertise and experience that may be utilised in various aspects of operational management. The Bank has developed internal environmental technology expertise through the environmental investment mandates. Moreover, Norges Bank’s climate risk assessments have resulted in divestment from companies the Bank deems to represent especially high long-term risk. These have primarily been small companies. The Government Pension Fund 2019 white paper describes the Bank’s operationalisation of the coal criterion, including data collection and contact with companies. The Bank has through the application of the coal criterion accumulated knowledge and data on individual companies’ climate risk management, including their strategies for emission reduction, technology development and climate policy adaptation, which may support and supplement the Bank’s application of other tools. In May 2020, the Bank announced exclusions of several companies on the basis of unacceptable greenhouse gas emissions and contribution to severe environmental damage; see the discussion in chapter 6.2 (in Norwegian only).

Norges Bank’s climate risk efforts have been evolving since 2006. The Ministry expects the Bank to continue to report comprehensively on its work on climate risk, including progress in attaining the objectives stipulated in the overall climate risk management strategy. Reporting should to the extent appropriate be in line with leading frameworks (TCFD), including climate scenarios for the portfolio and relevant sub-portfolios, relevant metrics and stress testing.