3 Investment strategy

3.1 Introduction

The Government Pension Fund was established in 2006, encompassing the former Government Petroleum Fund and National Insurance Scheme Fund. The purpose of the Government Pension Fund is to support government savings to finance the pension expenditure of the National Insurance Scheme and long-term considerations in the application of government petroleum revenues. To ensure that the petroleum revenues are contributing to the stable development of the Norwegian economy, the revenues shall be phased into the economy gradually, whilst the savings shall be invested outside Norway. The savings of the Pension Fund take the form of general fund accumulation. The Fund is fully integrated with the Fiscal Budget, in order to facilitate growth in the fund being a reflection of the State’s actual accumulation of financial assets. Consequently, there is no requirement that the capital of the Pension Fund should at all times equal a specific share of state pension liabilities under the National Insurance Scheme. This means that the Pension Fund is better placed to withstand short-term return fluctuations in the securities market than are many other funds.

Whilst the capital of the Government Pension Fund may only be spent once, the real return thereon may be spent every year without reducing its value. The fiscal rule for budget policy dictates that state petroleum revenues should be phased into the economy more or less in line with developments in the expected real return on the Government Pension Fund – Global, cf. Box 1.1. This contributes to maintaining the purchasing power of the Pension Fund over time, and enables a long time horizon to be adopted for the investments of the Fund.

There is a broad political consensus that the Pension Fund should be managed with a view to achieving the maximum possible return within a moderate level of risk. The Ministry of Finance has formulated a long-term investment strategy ensuring that the capital is invested in a broadly based portfolio comprising securities from many countries. The long investment horizon of the Fund means that the portions invested in various asset classes and geographical regions can be determined on the basis of assessments of expected long-term returns and risks.

The investment strategy chosen by the Ministry of Finance is a decisive influence on the Pension Fund’s expected return and risk, and reflects a trade-off between these two characteristics. The Pension Fund shall act as a financial investor, and not as a tool for exercising strategic ownership in individual companies. The Fund is characterised by good diversification of risk, as a result of it being invested in securities issued by many different states and by companies in many different countries. Priority is accorded to achieving broad political agreement as to the investment strategy of the Fund, and a high degree of openness as far as the management of its capital is concerned. This strengthens the credibility of, and confidence in, the Fund.

1 The Ministry of Finance has designed the benchmark portfolio of the Government Pension Fund – Global. In choosing the regional weights, one has taken into consideration, inter alia , Norway’s import weights. The composition of the benchmark portfolio within each region is determined by the inter-country market value weights. The benchmark portfolio of the Government Pension Fund – Norway is designed by the Executive Board of Folketrygdfondet within investment limits laid down by the Ministry. The investment limits are specified as a percentage of acquisition cost, whilst the benchmark portfolio is specified as a percentage of market values.

Source Ministry of Finance

The investment strategy of the Government Pension Fund is defined by the general investment limits and the benchmark portfolios of the Government Pension Fund – Global and the Government Pension Fund – Norway, respectively. The benchmark portfolios of the Government Pension Fund comprise equity and bond indices from different countries, cf. Figure 3.1. These indices include representative security samples, and developments in such indices reflect, in large part, market developments in the relevant countries. The return on the Government Pension Fund will, to a large extent, mirror market developments for the securities included in the benchmark portfolios. Calculations in Chapter 2 show that 90–95 pct. of the return on the Government Pension Fund – Global has been determined by the choices made by the Ministry of Finance when designing the benchmark portfolio, whilst the remaining 5–10 pct. has been determined by the investment choice made by Norges Bank within the guidelines laid down by the Ministry. A more detailed description of how a benchmark portfolio is designed can be found in Box 3.1.

Source Ministry of Finance

The Government Pension Fund is mainly invested in listed equities and bonds. Equity investments represent ownership interests in the production of goods and services, and the value of such investments will therefore reflect, inter alia , expectations as to the future profits of businesses. Bond investments involve the granting of a loan to the issuer, to be repaid to the bondholder together with a predetermined interest payment. As shown in Figure 3.1, the Government Pension Fund – Global holds all its investments abroad, and its return in international currency is the relevant measure of developments in the Fund’s international purchasing power. The Government Pension Fund – Norway is primarily invested domestically, and its return is measured in Norwegian kroner. Norges Bank and Folketrygdfondet (also known as the National Insurance Scheme Fund) seek to achieve a higher return than dictated by the benchmark portfolios of the Government Pension Fund – Global and the Government Pension Fund – Norway, respectively.

The Ministry of Finance has defined an investment universe for the Government Pension Fund that is wider in scope than the benchmark portfolios. At the same time, there has been established a limit as to how much the managers may deviate from the benchmark portfolio, in the form of an upper limit on tracking error, cf. Box 2.1. By making investments in securities that fall outside the scope of the benchmark portfolio, Norges Bank and Folketrygdfondet are exploiting their permitted tracking error leverage for purposes of achieving excess return. The risk assumed through active management has only to a limited extent increased the risk of the Fund beyond the level implied by the benchmark portfolios of the two parts of the Government Pension Fund, cf. Box 3.2. Analyses in Chapter 2 show that about 99 pct. of the fluctuations in the return on the Government Pension Fund – Global may be attributed to the chosen benchmark portfolio. For the Government Pension Fund – Norway, the variations in the return on the benchmark portfolio have explained about 93 pct. of the variations in the return on the Fund.

The Ministry of Finance conducts regular reviews of the Pension Fund’s investment strategy. These regular reviews examine any new information against the basis underpinning previous decisions, and particular weight is attached to assessing those choices with a material impact on expected returns and risks. This effort is based, inter alia , on expectations as to long-term returns and risks. In evolving the long-term, general investment strategy of the Government Pension Fund – Global, the Ministry of Finance is drawing on advice from, inter alia , Norges Bank and the Strategy Council appointed by the Ministry of Finance. The resources devoted to the Ministry of Finance’s involvement with the Government Pension Fund have been expanded in recent years, and responsibility for efforts relating to the Fund’s framework, its long-term investment strategy and the follow-up of Norges Bank and Folketrygdfondet, has since the autumn of 2006 been assigned to a designated Asset Management Department.

3.1.1 The strategy of the Government Pension Fund – Global

The Government Pension Fund – Global has grown rapidly since the Fund received its first capital allocation in 1996, cf. Figure 3.3. Over the last decade, the Fund has turned into one of the largest funds in the world. It is estimated, on the basis of projections in the National Budget for 2007, that its market value will increase by about 70 pct. towards 2010, to just over NOK 3,000 billion. The fund capital has significantly outgrown the levels envisaged for the first few years after the initial capital allocations were made.

Source Ministry of Finance

There has been a gradual evolvement in the investment strategy of the Government Pension Fund – Global. The Fund has invested in equities since 1998. In 2000, emerging markets were included in the benchmark portfolio for equities, whilst in 2002 the benchmark portfolio for fixed income was expanded through the inclusion of non-government-guaranteed bonds (i.e. corporate bonds and mortgage-backed bonds). In 2004, new Ethical Guidelines were laid down for the Government Pension Fund – Global. In 2006, the investment universe was further expanded by, inter alia , the abolition of special limits concerning the credit risk associated with bonds and concerning the duration range of the fixed-income portfolio. Furthermore, general authority was granted for the use of instruments characterised by an inherent affinity with permitted assets, hereunder commodity derivatives and mutual fund units, cf. Chapter 5.

The Government Pension Fund has been established as an integral part of Norwegian economic policy. The Global part of the Fund is intended for long-term use, with the real return thereon being spent over time. Over the Fund’s first decade, valuable experience has been garnered concerning investments in listed and frequently traded securities.

In the current Report, the Ministry presents several changes to the investment strategy. Sub-chapter 3.2 discusses the equity portion of the Government Pension Fund – Global, and outlines an increase therein from the current 40 pct. to 60 pct. At the same time, one intends to increase the number of companies invested in, by including small companies in the equity benchmark, cf. Sub-chapter 3.3. Moreover, one intends to change the regulation of recognised markets and currencies, cf. Sub-chapter 3.4. Moreover, Sub-chapter 3.5 provides a discussion of the status of the Ministry’s effort to evaluate the possible inclusion of real estate and infrastructure as a new asset class under the Government Pension Fund – Global.

The changes and evaluations discussed in Sub-chapters 3.3 and 3.5 should be seen in the context of an evolvement of the investment strategy, which involves assessing the scope for adding less liquid investments in the investment universe of a long-term investor like the Government Pension Fund. 1 However, the majority of the Fund’s investments will continue to be held in the form of listed equities and investment-grade bonds. In a letter of 20 October 2006, Norges Bank has recommended that the Government Pension Fund – Global be permitted to invest in unlisted equities as well. The letter is appended to the present Report. The Ministry will revert with an assessment of this issue later.

3.1.2 The strategy of the Government Pension Fund – Norway

Material changes were made to the investment strategy of the Government Pension Fund – Norway during the course of 2006. The guidelines governing the investments of the Government Pension Fund – Norway were previously in the form of a set of rules laid down by the Storting. In the Revised National Budget for 2006, the Ministry presented new Regulations relating to the management of the Government Pension Fund – Norway, to replace this set of rules, in line with the discussion in Proposition No. 2 (2005–2006) to the Odelsting, On the Act relating to the Government Pension Fund. The intention was to create, inter alia , a more uniform system for determining the guidelines governing the two parts of the Pension Fund. The Regulations were adopted on 15 December 2006, and entered into effect on 1 January 2007.

Until the end of last year, a major part of the capital of the Government Pension Fund – Norway was held in the form of sight deposits with the treasury. The sight deposit arrangement has contributed to meeting the funding needs of the treasury, without making Folketrygdfondet, as the manager of the Government Pension Fund – Norway, a large individual player in market-traded government securities. In the Revised National Budget for 2006 it was proposed, in connection with the effort to develop the new management guidelines, that the sight deposit arrangements be terminated. It was pointed out that such arrangements were of less importance now than upon their establishment in 1967, and that they might contribute to a less transparent presentation of the net financial assets of the State. At the same time, the Government proposed that NOK 10 billion of the sight deposits be repaid in order to strengthen long-term state ownership in the Norwegian business sector, in line with the Soria Moria Declaration. The termination of the sight deposit arrangement implied that NOK 101.8 billion of the capital of the Government Pension Fund – Norway was written down on 29 December 2006.

Pursuant to the Regulations of 15 December 2006, No. 1419, relating to the management of the Government Pension Fund – Norway, the capital shall mainly be invested in equities and primary capital certificates listed on a Norwegian exchange, or in bonds and commercial paper issued by Norwegian enterprises. The Fund’s limit on investments in equity instruments corresponds to 50 pct. of its capital under management (measured at acquisition cost). Within an overall limit of 20 pct. of its capital under management, up to 10 percentage points may be invested in equities listed on exchanges in Denmark, Sweden and Finland, and bonds and commercial paper issued by enterprises domiciled in the said countries, respectively. Folketrygdfondet’s authority to invest the capital of the Government Pension Fund – Norway in bonds issued in other Nordic countries was introduced with effect from 1 January 2007.

It follows from the investment guidelines set out in the Regulations that Folketrygdfondet may also use interest rate and currency derivatives in the management of its fixed-income portfolio, cf. Folketrygdfondet’s letter of 10 November 2006. In its letter, Folketrygdfondet referred to the fact that interest rate swaps entered into with the State in relation to the sight deposits were used to manage the interest rate risk of the Government Pension Fund – Norway, and that the termination of the sight deposit arrangements would almost halve the duration of the fixed-income portfolio, to about two years. In order for the Executive Board of Folketrygdfondet to be able to fix a duration of the benchmark portfolio for the fixed-income investments that would be more or less in line with its previous duration, it would require permission to make use of interest rate derivatives for management purposes, and the most relevant interest rate management instrument would be interest rate swaps entered into with other investors. As far as permission to make use of currency hedging instruments is concerned, Folketrygdfondet pointed out that this would enable it to separate the interest rate exposure from the foreign exchange risk, when managing bonds issued in other Nordic countries.

The Ministry is of the view that there are good reasons for permitting Folketrygdfondet to make use of interest rate and currency derivatives in managing the interest rate and foreign exchange risk of the Government Pension Fund – Norway. It was therefore indicated, in a letter of 5 December 2006 to Folketrygdfondet, that the Ministry would grant such permission in connection with the adoption of the Regulations relating to the management of the Government Pension Fund – Norway. In its letter to Folketrygdfondet, the Ministry also writes the following:

“In this context, the Ministry has attached considerable weight to Folketrygdfondet’s undertaking to refrain from investing in interest rate derivatives or currency hedging instruments until it has been demonstrated that there exist satisfactory risk systems and control procedures for the instruments to be used in the management of GPFN, hereunder that appropriate limits for counterparty risk have been established, cf. the letter of 10 November. This implies, inter alia, that Folketrygdfondet shall ensure that formal internal procedures within areas like trading processes, limit control and the follow-up of internal instructions in general, are in place and work as intended. The letter also states that excess liquidity resulting from the interest rate derivatives contracts shall be invested in the money market or in corporate bonds with a floating rate of interest, such as not to affect the equity investments. Furthermore, Folketrygdfondet writes that there will be imposed strict reporting requirements as far as concerns the distinction between ordinary securities and derivatives products, and their separate effects in terms of the return and risk of the overall portfolio.”

The termination of the sight deposit arrangements resulted in the equity portion increasing, in line with the investment limits, from 20 to just below 50 pct. of the capital under management (measured at acquisition cost). When taken in isolation, this increases the expected percentage return and the annual fluctuations in the return on Folketrygdfondet, since the equity portfolio now accounts for a larger portion of the overall portfolio than was previously the case, cf. the discussion in the Revised National Budget for 2006. The value of the equity portfolio of the Government Pension Fund – Norway was NOK 63.4 billion as per the end of last year. This represented an equity portion of 59.3 pct. (measured at market value). As per the end of last year, the investment limits defined for the Government Pension Fund – Norway would permit it to increase its equity investments by about NOK 5 billion. If this limit had been exploited to the full, the equity portion would have been about 65 pct. as per the end of last year.

3.2 The equity portion of the Government Pension Fund – Global

3.2.1 Previous assessments of the equity portion

The Regulations relating to the management of the Government Petroleum Fund were originally presented in the Revised National Budget for 1996. The Regulations dictated that the Petroleum Fund should be invested in government bonds, pursuant to more or less the same guidelines as applied to the foreign exchange reserves of Norges Bank.

Projections of government finances in the Long-Term Programme 1998-2001 (Report No. 4 (1996-97) to the Storting) indicated that the Fund would become larger than previously estimated, and that one would not need to make drawdowns on the Fund until later than previously assumed. In the Revised National Budget for 1997, the issue of whether to adopt a longer investment horizon therefore presented itself. Against this background, one considered the scope for expanding the investment alternatives to encompass equities as well. In its Report, the Jagland Government concluded as follows:

“Against this background, the Government will aim to open parts of the Petroleum Fund to investments in equity instruments. One intends to present new guidelines in the autumn of 1997, which will enter into effect as from 1 January 1998.

It is necessary to examine in more detail what portion of the Fund should be opened for investments in equities. There is some variation amongst long-term investors internationally as to what portion of their portfolio is comprised of equities. An equity portion in the 30-70 pct. range would appear to be common amongst this type of investors. The Ministry’s preliminary assessments suggest that the equity portion of the Petroleum Fund’s portfolio should at this stage be in the region of 30-50 pct. The issue will be examined more closely when preparing new guidelines.”

It was pointed out in the Revised National Budget for 1997 that such an assessment had to involve a trade-off between expected return and risk. It was stipulated that the objective pursued in managing the Fund should, in principle, be to invest its capital in such a way as to maximise the international purchasing power of the Fund around the time when it was likely that one would have to make drawdowns on the Fund, subject to an acceptable level of risk. Moreover, it was pointed out that variations in the return on the Fund from one year to the next are of less importance.

A majority of the members of the Storting’s Standing Committee on Finance and Economic Affairs supported the principles underpinning the new guidelines for the management of the Government Petroleum Fund, as outlined in the Revised National Budget for 1997. New Regulations relating to the management of the Petroleum Fund were presented to the Storting, in the autumn of 1997, in the National Budget for 1998. The equity portion of the benchmark portfolio was fixed at 40 pct. At the same time, there was defined a permitted range of 30 – 50 pct. for the equity portion of the Fund.

Furthermore, the National Budget for 1998 discussed, inter alia , whether this was the appropriate time to phase equities into the Fund, given the fact that the preceding years had witnessed a strong upturn on a number of the large stock exchanges internationally. The Report went on to state that:

“One cannot rule out the possibility that the long-term cyclical upturn may be followed by a cyclical downturn in equity prices. However, based on the premise that all available information is reflected in equity prices, the lengthy upturn on certain exchanges cannot in itself provide any information as to whether now is a good or bad time to invest. Historical experience suggests that the timing of investments is of less importance if one adopts a long investment horizon than if one operates with a short investment horizon. When entering the stock market through a number of investments spread over a long period of time, one effectively spreads the risk associated with whether one is entering the market at times that will in retrospect be perceived as favourable or unfavourable. In addition, there is the fact that the Petroleum Fund will be growing in coming years. If one was to postpone the phase-in of equities in the portfolio, this would result in one having to purchase correspondingly more equities later, without any assurance that such would be a better time. Consequently, long-term considerations may suggest that one should expand the equity portfolio of the Petroleum Fund gradually in line with increases in the Fund’s size.”

The last occasion on which the equity portion was examined was in the National Budget for 2004. The Report referred to the major fluctuations in the stock market since the Fund started investing in equities. It was emphasised that the factors taken into consideration when deciding to include equities in 1997, had not changed, and that the purpose of the Fund suggested that one should not attach much weight to short-term fluctuations in its return. The Report goes on to state that:

“In its letter of 28 August this year, Norges Bank concludes that the premises on which the decision was based have not changed significantly, cf. above. The Ministry agrees with this assessment. Although the return fluctuations have been strong, given the information available in 1997, the return has not fallen outside the opportunity set outlined as a basis for the decision to adopt a 30-50 pct. equity portion.

The previous occasion on which the equity portion of the Petroleum Fund was discussed was in the National Budget for 2002. The recommendation was then to examine this in a broader perspective. The Ministry is of the view that the current equity portion of 30-50 pct. comes close to what is perceived as an acceptable level of risk for the Fund. Although a higher equity portion will result in a higher expected return, one must when increasing such portion also accept that the risk of larger variations in such return will increase.

Based on an overall assessment, the Ministry does not consider it imperative to increase the equity portion at the present point of time. One therefore intends to maintain the current 30-50 pct. portion. Within this range, Norges Bank may change the equity portion if it deems it appropriate to do so, based on an assessment of market developments, whilst at the same time keeping within its overall risk limit.”

3.2.2 The recommendations of Norges Bank and the Strategy Council

The fund capital has grown well beyond the level envisaged when the equity portion was fixed in 1997, as well as the level envisaged when the equity portion was last reviewed in the National Budget for 2004. Nevertheless, the basis underpinning the investment choices of the Government Pension Fund remains the same. There is broad political agreement that the level of risk should be moderate, but also that one should seek to maximise returns. Like on previous occasions, the assessment is premised on a long time horizon for the investments. Whilst in 1997 it was envisaged that it would become necessary to make drawdowns that would reduce the real value of the Fund, the budget policy guidelines introduced in 2001 suggest that spending over time will match the real return on the Fund, cf. Box 1.1. This supports the view that the assessment of what constitutes an acceptable level of risk should emphasise the fluctuations in the return on the Fund over periods of many years, and that developments from one year to the next are of lesser importance. Like with the previous assessments, it is also now relevant to examine the investment choices facing the Government Pension Fund – Global from the perspective of the choices made by other large funds.

There has throughout been broad political agreement as to the Fund’s investment strategy. Nevertheless, when the equity portion of the Fund was fixed at 40 pct. in 1997, some parties in the Storting did argue against investing in equities. At that time, the counter-arguments were centred, in particular, on the absence of ethical guidelines and on concern about the increased level of risk assumed by the Fund. The Ministry is of the view that there are several reasons why the assessment as to what constitutes an acceptable level of risk on the part of the Fund may now yield a different conclusion. Experience and knowledge gained from investments in equities may have changed perceptions as to what constitutes an acceptable level of risk for the Fund’s investments. The fiscal policy guidelines are designed in such a manner that major changes in the fund capital have no direct fiscal policy impact in the short run, and the experience with equity investments since 1998 has demonstrated, inter alia , that one is able to handle major fluctuations in returns. Furthermore, ethical guidelines for the investments of the Fund have been adopted. The ethical underpinnings of the Fund are furthered through exclusion of individual companies, and through the promotion of corporate governance in those companies in which the Fund is invested. For purposes of a renewed assessment of the equity portion, it is also of relevance that one has gained, since the fund accumulation was initiated, considerable expertise as far as equity management is concerned, together with experience in the handling of fluctuating portfolio performance from one year to the next.

The portion of the Fund to be invested in equities is the single most important decision in determining the Fund’s risk taking. Consequently, deliberation of the equity portion of the Fund is a key feature of the Ministry’s efforts relating to the investment strategy of the Fund. In the Revised National Budget for 2006, the Ministry announced that one would be reverting with a new assessment of the Fund’s equity portion, and that the review thereof would be based on, inter alia , input from Norges Bank and the Strategy Council.

In the National Budget for 2007, the Ministry of Finance discussed the two recommendations it had by then received from Norges Bank and the Strategy Council, to the effect that the equity portion of the Fund should be increased. The recommendations were made public at the same time, and are available on the Ministry’s website (www.finansdepartementet.no).

In a letter of 10 February 2006, Norges Bank wrote to the Ministry of Finance that:

“Norges Bank is of the view that an overall assessment of historical evidence and our present market perception suggests that the percentage of equities held by the Government Pension Fund – Global should be increased. The Ministry of Finance should weigh the expected gain from a 50 or 60-percent allocation to equities against the increase in risk”.

In a letter of 2 June 2006, the Strategy Council wrote to the Ministry of Finance that:

“Based on Norges Bank’s own analyses and model computations, as well as our own evaluations and reviews of relevant financial literature, the Council recommends that the equity allocation allotted to the benchmark portfolio of the GPFG be increased to 60 percent”.

Reference is made to the discussion in the National Budget for 2007, for a more detailed explanation of the two recommendations from Norges Bank and the Strategy Council. In this context the Ministry wrote, inter alia , that:

“The recommendations are based on extensive supporting documentation, and the Ministry wishes to examine this important issue in more detail to ensure that all aspects of the matter have been sufficiently elucidated.”

Based on its advisory agreement with the Ministry of Finance, Norges Bank carried out a new review of the investment strategy of the Government Pension Fund – Global in 2006. The analysis is documented in Norges Bank’s Strategy Report for the Government Pension Fund – Global and in a letter of 20 October 2006 from Norges Bank to the Ministry of Finance. In its letter, Norges Bank reiterates, inter alia , its recommendation that the equity portion of the Fund should be increased:

“In our letter of 10 February 2006, Norges Bank recommended that the Ministry increase the equity portion in the benchmark portfolio of the Government Pension Fund – Global. The updated analysis we have now completed confirms the findings that underpinned that recommendation: A higher equity portion will increase both expected return and return volatility. The trade-off between expected return and volatility appears to be attractive in the long term. The risk of a negative accumulated real return increases slightly, but it is highly likely that an equity portion will be profitable. If we look at a conditional probability distribution, we find that the expected loss, if an increased equity portion turns out not to be profitable, is relatively modest.”

The letter from Norges Bank is appended to the present Report, and is available at www.finansdepartementet.no.

3.2.3 The Ministry’s assessment

There is no definite answer to the question of what is the optimal equity portion. Instead, the choice of equity portion needs to be based on a trade-off between the expected return and risk associated with various investment choices. As far as the Pension Fund is concerned, such a trade-off is ultimately a political decision. The Ministry of Finance has gathered advice, where the expected return and risk associated with various investment choices have been explained and calculated. Calculations of consequences in terms of the returns and risks of the Pension Fund have been performed on the basis of historical returns going back to the year 1900. In addition, model computations have been carried out regarding the expected consequences over the coming 15-year period, based on other assumptions than those implied by historical returns, cf. Boxes 3.3 and 3.4. Moreover, experience with the Fund since 1998, when the Fund made its first investments in equities, is of relevance to the trade-off between expected return and risk. From a historical perspective, fluctuations in the stock market have been large during parts of this period. This is illustrated in Figure 3.4. In the Figure, the years since 1998 are framed. It may be noted that 1999 was one of the best years in the stock market over the last 107 years, and that 2002 was the second worst year since 1900. Nevertheless, the experience has been that these fluctuations could be handled. A broad political foundation underpinning the chosen equity portion, and fiscal policy guidelines that offered the flexibility required to handle such situations, were important in this context.

Source Ministry of Finance/Dimson, Marsh and Staunton

Textbox 3.4 Model computations

The effect of increasing the equity portion on the expected return on, and risk of, the Fund may be analysed by way of model computations (simulations). Such computations are based, inter alia , on assumptions as to how returns and risks develop within different asset classes over time, and the covariation between them. One may, on the basis thereof, simulate developments in the value of the fund capital until a selected future date, cf. Figure 3.5.¹ Figure 3.5 illustrates how developments in the value of the Fund may be stimulated by “drawing” several thousand different outcomes over 15 years. The yellow line represents the typical outcome (the median). The Figure illustrates that some of the outcomes will be considerably better/weaker than the representative outcome.

Source Ministry of Finance

Norges Bank refers to such model computations in its letters of 10 February and 20 October 2006. The main assumptions on which the analyses are based are as follows:

• Equities are expected to render an annual return over time that is about 2 percentage points higher than on bonds (i.e. that the risk premium for equities is 2 pct.).

• A weak degree of “mean reversion” is assumed as far as equity returns are assumed. The assumption that returns will over time revert to long-term “normal levels” or equilibrium levels is based on analyses of historical returns in the US. Such an assumption implies that the risk associated with equity investments is somewhat lower than it would otherwise have been when the investment horizon is extended. However, this assumption is not of decisive importance as far as the recommendation to increase the equity portion is concerned.

• The model simulations are based on rates of return following a normal distribution. It is known that the distributions will in practise be somewhat more skewed towards the tail ends than suggested by the normal distribution, thus implying that the probabilities of extreme outcomes, in both negative and positive terms, are underestimated in the simulations.

However, the assumptions underpinning the respective analyses deviate somewhat as far as risk estimates are concerned. In the letter of 10 February from Norges Bank, which is also discussed in the National Budget for 2007, the calculations are made under risk assumptions based on developments since 1900. In the Strategy Report of 29 November 2006, corresponding model computations are made on the basis of three alternative scenarios. In the scenario deemed most likely by Norges Bank, the risk estimates are based on developments over the last twenty years. A deflation scenario and a stagflation scenario have also been prepared, cf. Table 3.1. The covariation between

equities and bonds is assumed to be very low in the main scenario and in the deflation scenario. However, the stagflation scenario assumes a high covariation on the basis of experience from the 1970s. Norges Bank has utilised a currency basket based on expected import weights.

Table 3.1 Annual nominal returns (geometric) and risks measured in the currency of each country/region. Percent

		Main scenario		Deflation		Stagflation
Average global inflation rate 2006 –21		2.1		0.2		5.5
		Return	Standard deviation	Return	Standard deviation	Return	Standard deviation
Bonds:	Americas	5.4	4	3.5	4	7.4	9
	Japan	2.9	4	1.1	4	4.8	9
	Asia/Oceania ex Japan	5.2	5	3.3	5	7.2	10
	UK	4.7	5	3.2	5	7.1	10
	Europe ex UK	4.2	3	2.6	3	6.4	8
Equities:	Americas	6.6	15	2.4	17	5.6	17
	Japan	4.7	20	1.4	22	3.8	22
	Asia/Oceania ex Japan	6.6	21	2.4	23	5.6	23
	UK	7.6	16	3.1	18	6.8	18
	Europe ex UK	5.7	18	1.6	20	5.0	20

Source Norges Bank

The model computations concerning how an increase in the equity portion from 40 to 60 pct. may influence the expected return on, and risk of, the Fund are illustrated in Figure 3.6.

Source Norges Bank and the Ministry of Finance

The Figure may be read to state that in about 25 pct. of the cases, the accumulated real return is 45 pct. or less irrespective of whether the equity portion is 40 pct. or 60 pct. The conclusions from the model computations may be summarised under reference to Items 1-3 of Figure 3.7:

1. There is a high probability that a 60 pct. equity portion will generate a higher accumulated return after 15 years. In some cases the return is considerably higher than in case of a 40 pct. equity portion.

2. There is a probability of about 25 pct. that a 60 pct. equity portion will generate a lower aggregate return than a 40 pct. equity portion. Even for these outcomes (all observations to the left of the vertical line), the black line is not significantly higher than the orange one. This means that the reduction in return is limited, even in the 25 pct. or so least favourable outcomes.

3. The probability that the portfolio will generate a negative accumulated real return after 15 years is low, irrespective of whether the equity portion is 40 pct. or 60 pct.

¹ The simulation is based on a common assumption to the effect that the prices of the various asset classes follow a log-normal distribution, which ensures that asset prices cannot become negative. This implies an assumption to the effect that the return follows a normal distribution. The price processes that determine developments in the value of the asset classes over time are described by so-called geometric Brownian motions. For each asset class and exchange rate, numbers are randomly selected from a normal distribution, where the numbers represent the shocks (the surprises) influencing price developments relative to trend. This is repeated step-by-step until a chosen future date (time horizon). This generates possible “paths” for the asset price, and a corresponding portfolio return in a selected benchmark currency. This procedure is repeated a large number of times (typically 6,000 times). This generates a distribution of possible portfolio returns, which is used for studying the statistical qualities of the portfolio.

Figure 3.7 shows the annual real return on a portfolio of long-term bonds since 1900. The Figure shows that returns on bonds are less variable than those on equities. This is reflected in the fact that a much higher number of years is clustered into the two midmost bars, where returns are close to the historical average. However, the number of years with negative returns is much higher. As much as 40 of the last 107 years have registered a negative real return on bonds, as compared to 29 of the last 107 years for equities. Returns for the years 1998–2006 have been framed, and show, inter alia , that the Fund has since 1998 experienced some of the best years in the bond market since 1900. Nevertheless, the Fund has since 1998 made about NOK 100 billion from investing 40 pct. of the capital in equities rather than fixed income.

Source Ministry of Finance/Dimson, Marsh and Staunton

Consequences in terms of expected return

Expectations for a higher average return on equities than on bonds are based on both historical experience over more than 100 years and on economic theory. Investors will not be willing to assume more risk without being compensated in the form of higher expected return. Consequently, investors will not been purchasing equities until prices indicate that the expected return will be sufficiently high to compensate for the difference in risk, cf. Box 2.8. However, the magnitude of this risk premium is more uncertain. Based on the assumptions adopted in the main alternative presented by Norges Bank, as explained in more detail in Box 3.4, the effect on the Fund’s return, as measured in NOK, may be illustrated by the following calculation:

• With a fund valued at NOK 1,800 billion, the expected increase in value from increasing the equity portion from 40 pct. to 60 pct is estimated at about NOK 190 billion after 15 years. If the equity portion is instead increased to 50 or 80 pct., the increase in value is estimated at NOK 100 and 340 billion, respectively.

• There will be new capital inflows to the Fund as the result of government petroleum revenues over the coming 15 years. If the return on the estimated inflow of new capital is included in the computations, the expected increase in value from investing 60 pct. in equities instead of the present 40 pct. expands to about NOK 240 billion. With an equity portion of 50 or 80 pct., the expected increase in value will in such case be NOK 130 and 450 billion, respectively.

The above estimate is based on uncertain assumptions. At the same time, there are several reasons for believing that these estimates are moderate. The calculations assume, inter alia , that equities offer a risk premium in the form of an average annual excess return of about 2 percentage points relative to bonds. 2 Consequently, increasing the equity portion from 40 to 60 pct. will increase the expected annual return on the Government Pension Fund – Global by about 0.4 percentage points. This expected risk premium on equities is less than half of what the excess return has been over the last 100 years, and about half a percentage point less than what many other funds assume. However, the Ministry acknowledges that there are several reasons why one should not assume that historical experience is representative of what may be expected in future. Such caution is also in line with the consensus view amongst large investors, cf. Box 3.3. Furthermore, the estimates for the effects on the Fund’s return are expressed by way of the most typical outcome (i.e. the outcome at the middle of the set of possible outcomes), and not by the average outcome. This implies that the computations have attached less weight to extremely favourable outcomes that would have a low probability of occurring.

Consequences in terms of risk

Model computations showing how the returns and risks of the Fund are changed, when changing the equity portion, are presented in Tables 3.2 and 3.3. Column three shows that the fluctuations in returns from one year to the next, as measured by the standard deviation of annual returns, increase from 6.2 pct. to 8.7 pct. Given the long time horizon of the Fund, the uncertainty associated with the average return over 15-year periods is more relevant. Risk as measured in this manner increases from 1.6 pct. to 2.2 pct. if the equity portion is expanded to 60 pct. An average return of 4 pct., and an accompanying standard deviation of the average return over 15-year periods of 2.2 pct., may here be interpreted to mean that the average annual real return will in two out of three cases fall within the range of 1.8 – 6.2 pct. over a 15-year period.

Besides, the risk of overall loss over a longer time period increases somewhat when the equity portion is increased. Nevertheless, the model computations from Norges Bank show that the probability of loss over 15-year periods will remain relatively low even if the equity portion is increased, cf. column five of Table 3.2. With 60 pct. held in equities, the probability of a negative accumulated return after 15 years is estimated at 3.7 pct. The same model computations show that the expected negative excess return, assuming that increasing the equity portion turns out to be unprofitable, will be relatively modest.

The computations in Table 3.2 are based on the main alternative from the Norges Bank analyses, cf. Box 3.4. The assessment of risk in these computations is based on the assumption that the experience from the last 20 years is of more relevance to the assessment of the outlook for the coming 15-year period than is the period 1900–2006 as a whole. This results in, inter alia , a lower expected risk associated with bonds than was assumed in the computations discussed in the National Budget for 2007. Norges Bank takes the view that those assumptions attributed too high a level of risk to bonds. Nevertheless, the increase in the risk level of the Fund associated with expanding the equity portion, as estimated on the basis of the assumptions adopted in the discussion in the National Budget for 2007, is presented in Table 3.3.

Comparing Tables 3.2 and 3.3 shows how changed assumptions influence the model computations of return and risk. In general, the overall risk level of the Fund, with both 40 and 60 pct. equities, is lower under the updated assumptions. For example, the risk as expressed by way of the probability of a negative accumulated aggregate return after 15 years is 3.7 pct. under the updated assumptions, as compared to 6.3 pct. under the assumptions adopted in the discussion in the National Budget for 2007. Nevertheless, both computations show a low probability of a negative accumulated return.

The National Budget for 2007 stated that the estimated probability of obtaining a lower return after 15 years with a 60 pct. equity portion, instead of a 40 pct. equity portion, was about 25 pct. Under the updated assumptions, this probability estimate remains at about 25 pct. The risk as expressed by way of the amount of the expected negative excess return in the 25 pct. least favourable cases, is now calculated at about 10 pct. on an accumulated basis, or 0.6 pct. per year. This is somewhat higher than the accumulated negative excess return as the result of the original assumptions, which was in the region of 6 pct. On the other hand, the expected aggregate return on the Fund with 60 pct. equities is now a small positive figure in even the 25 pct. least favourable cases.

The model computations are likely to somewhat underestimate the probability of loss, cf. Box 3.4. The model estimates as to how the probability of a negative accumulated return is affected by increasing the equity portion are therefore, as pointed out by Norges Bank, more reliable than the estimates as to the magnitude of such losses.

The increase in the risk of accumulated loss over a longer period of time, based on historical developments, is illustrated in Figure 3.8. There are only seven distinct 15-year periods since 1900. The Figure is therefore based on rolling 15-year periods that show the historical real return on the portfolio with a 60 pct. and 40 pct. equity portion, respectively. The portfolio featuring 60 pct. equities then turns out to be more profitable than the portfolio featuring 40 pct. equities over 87 of the 93 rolling 15-year periods. For the remaining six periods, the negative excess return from holding 60 pct. equities was limited. The least favourable 15-year period for equities according to these computations is the 15-year period up to and including 2004. Over this period, the portfolio comprising 60 pct. equities would have generated an accumulated return that was 15 percentage points lower than one comprising 40 pct. equities. The best 15-year period for the portfolio holding the highest equity portion was the period until 1963, when the portfolio comprising 60 pct. equities would have delivered an accumulated return that was 182 percentage points higher than the portfolio comprising 40 pct. equities. At the same time, the statistical robustness of these computations is weak, because the 15-year periods are overlapping each other in such a manner that the return from one year may be included in up to 15 of the 93 periods. Such overlapping observations therefore tend to underestimate the probability of loss.

Source Ministry of Finance/Dimson, Marsh and Staunton

Both historical returns and the model computations show that a higher equity portion results in more pronounced fluctuations in the return on the Fund from one year to the next, and in a somewhat higher probability of a negative accumulated return over 15 years. However, the Ministry is of the view that this increased risk remains moderate and that it can be accommodated, at least with an equity portion of 60 pct.

Given the very long investment horizon, the possibility of extreme events – e.g. relating to wars, natural disasters, climate changes and major financial breakdowns – is of relevance to the assessment of the Fund’s risk. History shows that equity investments have been remarkably robust in the face of a number of major crises. Figure 3.9 shows that events like, for example, world wars, stock market crashes, etc., have in a longer time perspective only had a moderate impact on equity returns in the markets where the Fund is invested. The real value of the investment has recuperated to its level before the crisis occurred after only a few years.

¹ The return is calculated on the basis of portfolio weights that are almost the same as the weights of the Government Pension Fund – Global. A logarithmic scale has been adopted. This means that losses and gains that are of the same percentage magnitude will be shown as equally large, irrespective of when these occurred over the time period covered.

Source Ministry of Finance/Dimson, Marsh and Staunton

One might envisage a number of alternative “crises” that may occur over the coming century. Problems relating to greenhouse emissions are an example. It is obvious that climate-related problems may impact on the returns on financial investments. At the same time, it must be assumed that market prices are already reflecting publicly available information concerning the possibility of such a crisis. An “environmental crisis” would have different effects on different companies. For some companies, this may present a profit opportunity. For others, it will induce losses. It is difficult to predict how this will affect the stock market as a whole. Reduced economic growth may reduce corporate earnings. At the same time, the risk premium may increase, which would reduce returns in the short run, whilst increasing expected returns, as compared to those on bonds, in the longer run. The best strategy for reducing the risk to the Government Pension Fund from such developments is therefore to spread the investments as much as possible, across different companies, industries and countries. In addition, one may exercise the Fund’s ownership influence to change corporate attitudes to the environment.

As far as the returns on bonds are concerned, the key issue is how any climate problems will affect economic growth. The average duration of the bonds in the Government Pension Fund – Global is about five to six years, cf. Box 3.2. The capital currently invested in bonds therefore needs to be reinvested several times in coming decades. If climate problems result in reduced economic growth, this may also mean lower real returns on the bonds purchased by the Fund. Consequently, bonds are not risk-free in such a situation either.

Assessment of the equity portion in a broader perspective

In addition to the model computations and the analysis of historical risk, the Ministry has attached weight to certain more general considerations regarding the risk associated with the Fund’s investments, which suggest that the risk resulting from an equity portion of 60 pct. may, in the view of the Ministry, be characterised as moderate:

• The overall value of the petroleum wealth comprises remaining oil and gas reserves and the current pension fund. By selling oil and gas, and investing in equities and fixed income, the petroleum wealth has been spread across more assets, and the overall risk has been reduced.

• The return on long-term investments in real interest rate bonds is now significantly lower than when the current investment strategy was established in 1997. This means that over this period it has become more challenging to achieve a real return of 4 pct. over time with the current equity portion of 40 pct.

• When assessing risk over very long time horizons, it is also an important consideration that equities and bonds are two fundamentally different investments. By purchasing equities, we purchase ownership interests in the world’s production capacity . By purchasing bonds we are extending a nominal loan to businesses and governments, as it is not possible to invest the entire Fund in real interest rate bonds, because of the Fund’s size. By investing in nominal bonds we know, with a high degree of certainty, what the nominal return will be over the next few years. Nevertheless, such investments have been risky in the longer run, and the Strategy Council points out, in its letter to the Ministry of Finance, that the risk associated with bonds has increased with longer investment horizons. This has partly to do with the fact that the last 100 years included periods of very high inflation, which have resulted in negative real returns.

It is also relevant, for purposes of checking the Ministry’s assessments, to look at what decisions other large funds have made in their choice of equity portion. A comparison with large pension and reserve funds shows that, whilst the Government Pension Fund – Global holds a bond portion of 60 pct., other funds more commonly feature a bond portion of 30 pct. Generally speaking, other funds also hold a higher equity portion than does the Government Pension Fund – Global. Large pension funds in Europe and the US commonly hold an equity portion in excess of 50 pct., whilst large government reserve funds commonly hold an equity portion of about 60 pct., Table 3.4 in Box 3.5.

Amendments to the guidelines

In the guidelines for the management of the Government Pension Fund – Global, it is stipulated that the Fund’s strategic benchmark portfolio comprises 40 pct. equities and 60 pct. fixed income. The permitted range for the actual equity portion is 30 – 50 pct. There are two reasons for maintaining a corresponding range when increasing the equity portion.

• The stock and bond markets in the various geographical regions develop differently from one month to the next. In line with this, the portfolio’s allocation between equities and fixed income will also move away from an equity portion of 40 pct. In order to avoid unnecessary transaction costs, the monthly inflows of capital are used to purchase equities or bonds in those regions that have shown the weakest development over the last month. To further reduce the need for transactions, it is permissible for the Fund’s actual benchmark portfolio after the capital inflow to feature an equity portion that deviates somewhat from the strategic portion of 40 pct. over short periods.

• Another reason for permitting a range around the 40 pct. equity portion is the possibility that the manager may, based on a perception of the market situation, wish to over or underweight equities relative to fixed income, as compared to their distribution in the Fund’s actual benchmark portfolio. Such deviations from the benchmark portfolio must take place within the tracking error limit, which has been fixed at 1.5 pct. by the Ministry. Thus far, Norges Bank has only to a limited extent chosen to over or underweight equities relative to fixed income, as can also be seen from the analysis of the excess returns in Chapter 2.

The Ministry is of the view that it would be appropriate, when increasing the equity portion, to maintain the current range for the permitted equity portion of +/- 10 percentage points relative to the equity portion of the strategic benchmark portfolio. A higher equity portion in the Fund will, ceteris paribus , somewhat increase the need for such a range. At the same time, the current guidelines for rebalancing and the large expected inflows to the Fund will, in combination with Norges Bank’s strategy for achieving excess return, all contribute to preventing the equity portion held by the Fund from becoming significantly higher than the one featured by the strategic benchmark portfolio over time. One indication of this is the Fund’s actual equity portion as per the end of each quarter since 2002, which has never been higher than 42.6 pct. (December 2003) or lower than 36.3 pct. (September 2002).

Conclusion

The overarching investment choices are of decisive importance to the expected return and risk of the Fund. The trade-off between risk and expected return is a political decision. Extensive and thorough documentation to underpin such a decision has now been gathered through the Ministry’s own efforts and the advice obtained from Norges Bank and the Strategy Council. The Government has concluded, based on an overall assessment of such documentation, that an increase in the equity portion of the benchmark portfolio of the Government Pension Fund – Global from 40 to 60 pct. represents a good trade-off between the expected return and risk associated with the investments of the Fund. It is therefore proposed that the guidelines for the management of the Government Pension Fund – Global be amended, by way of the equity portion of the strategic benchmark portfolio of the Fund being fixed at 60 pct. At the same time, an equity portion in the range 50–70 pct. will be permitted. This implies that the bond portion of the strategic benchmark portfolio will be 40 pct. Consequently, the permitted range of the bond portion will be 30–50 pct.

Estimates as to the expected return are subject to considerable uncertainty, and historical experience shows that the return may vary a great deal. This uncertainty is so pronounced that the current estimate of 4 pct. has to encompass expectations of average real returns of 3, 4 as well as 5 pct. An increase in the equity portion does not justify an increase in the expected real return on which the economic policy guidelines are premised. Instead, the increase in the equity portion will make it more likely that a 4 pct. real return will be achieved over time. If the return turns out to be higher than expected, the amount that can be transferred to the Fiscal Budget will nevertheless be higher, because one can over time spend 4 pct. of a larger fund capital than would have otherwise been the case. Only if the achieved return over many years turns out to be significantly higher or significantly lower than 4 pct. would it be appropriate to assess whether the estimate of 4 pct. represents too high or too low an expectation as far as future returns are concerned.

The assessment of the consequences of increasing the equity portion is primarily focused on how this may influence the portfolio’s return and risk in the long run. The assumption as to the expected return in the long run is based, inter alia , on analyses of the market’s pricing of equities and bonds. At the same time, there is considerable uncertainty associated with market developments in the short run. It must therefore be acknowledged that the timing of the increase in the equity portion may in retrospect be perceived as more or less favourable. This risk is somewhat alleviated by the fact that the Fund’s investments will be spread out in time, in line with the expected growth in the Fund. Besides, the Ministry will attach considerable weight to ensuring that the adjustment to a higher equity portion is implemented with low transaction costs. This implies that the adjustment may take several years.

3.3 Broadening of the benchmark portfolio of the Government Pension Fund – Global to include small-cap equities

3.3.1 Introduction

At present, shares of listed companies with low capitalisation (small-cap equities) are included in the investment universe of the Government Pension Fund – Global, but are not included in the Fund’s benchmark portfolio. Equities in small listed companies form the largest individual segment amongst those listed markets that fall outside the scope of the benchmark portfolio.

The issue of investments in small-cap equities has been examined before. In a letter of 1 April 2003, Norges Bank recommended that small companies be included in the benchmark portfolio. Norges Bank pointed out that such a broadening would provide a better representation of the investment universe, and that the size and growth of the Fund calls for broad exposure to this part of the market as well. The Ministry of Finance deliberated the issue in connection with the National Budget for 2004, and concluded that one would not be including small companies in the benchmark portfolio at that point of time. The Ministry pointed out, inter alia , that the Graver Committee (NOU 2003: 22 Green Paper) had recently submitted its proposal on Ethical Guidelines for the Fund, and that one deemed it preferable to await the deliberation thereof before embarking on a significant expansion in the number of companies in the benchmark portfolio for equities.

In accordance with its advisory agreement with the Ministry of Finance, Norges Bank has again examined the issue of whether small, listed companies should be included in the benchmark portfolio. In its letter of 20 October, Norges Bank writes, inter alia , that:

“Small-cap equities make up a substantial segment of the market. It is difficult to see why the Fund, as a large and long-term investor, should have an exposure to this segment which is substantially lower than that of the market in general. There are also moderate diversification gains to be had from including these equities. If small-cap equities are added to the benchmark portfolio, higher returns can be expected without a significant increase in volatility in the portfolio.”

In a letter of 20 February 2007, the Strategy Council has also recommended that the benchmark portfolio be expanded to encompass the small-cap equity segment. The Strategy Council writes, inter alia , that:

“Based on considerations relating to the overall return and risk of the Government Pension Fund – Global, the Strategy Council recommends that the Fund’s benchmark portfolio for equities be expanded by inclusion of the small-cap equity segment. Although costs, when taken in isolation, will increase somewhat after such a broadening, it is likely that this will be covered by way of a better risk-adjusted expected return for the portfolio. Such a broadening will make the Fund’s benchmark portfolio more representative of developments in the international stock markets. Furthermore, an expanded benchmark portfolio will offer a more appropriate basis for assessing the active management of the Fund. The Council deems the proposed broadening to be a reasonable consequence of the Fund’s general investment strategy, which is to purchase a representative portfolio of the world’s stock market.”

3.3.2 The recommendations of Norges Bank and the Strategy Council

In its letter of 20 October, Norges Bank points out that small-cap equities may be added to the benchmark portfolio by way of the current benchmark index for equities, the FTSE All-World, which as per yearend 2006 comprised in excess of 2,400 large and medium-sized companies, being replaced by the FTSE All-Cap, which in addition contains approximately the 12 pct. smallest companies in each region as measured by market capitalisation. This means that the number of companies in the benchmark portfolio is increased to about 7,000 companies. The average capitalisation of the new companies is about 7 billion Norwegian kroner in the Americas and Europe, and about NOK 2 billion in Asia. In comparison, the average capitalisation of large and medium-sized companies in New Zealand, which is one of the smallest markets in the current benchmark portfolio, is about NOK 7 billion. Norges Bank points out that the benchmark portfolio of the Pension Fund will, following such a broadening, represent about 96 pct. of the stock markets included in the FTSE index, as compared to the current 85 pct.

Norges Bank writes that the main rationale behind the proposal is that small-cap equities form a large market segment that is excluded from the current benchmark portfolio, and that a broadening of the benchmark will give an exposure to this segment that is more in line with the market average. The Bank also attaches weight to the following arguments in favour of the benchmark index for equities being broadened to include small-cap equities as well:

• Historical returns from many countries show that small-cap equities have on average generated higher returns than the large company equities. The first studies to present this finding were published in the 1980s and in the early 1990s. Figure 3.10 illustrates that the observed excess return, also termed the “small-cap premium”, varies from one period to another. The red bars in the Figure show average historical excess returns per month for small equities, as calculated in the first studies for each country. The black bars show the average premium for the period following the publication of the first findings, whilst the blue bars show average monthly excess returns on small companies for the period after 2000. The figures are not comparable across countries, because of differences in time periods and in the definitions of which companies are “small”. Nevertheless, the Figure illustrates that returns on small-cap equities have been higher and lower than those on large-cap equities for periods of many years. For the markets with the longest time series, i.e. the United States and the United Kingdom, returns have been higher for small companies than for large ones throughout the period. 3

• Returns on small-cap equities have varied more than returns on large-cap equities. Nevertheless, the risk level of broad equity portfolios comprising both large and small companies has not been significantly higher than that of portfolios including only large companies. This is because movements in the return on small companies have not been perfectly aligned with movements in the return on large companies.

• When the first studies of the return on small companies were made public, the prevailing theoretical interpretation of the functioning of the markets was that investors could only expect to be compensated for assuming one particular risk factor in the stock market, which was the risk associated with holding a well-diversified portfolio of equities. The observations of higher historical returns on small-cap equities have therefore been labelled the best-documented deviation from the assumption of efficiently priced stock markets. Subsequently, one has sought to explain the difference in returns between small and large companies by way of multi-factor models, where the size of the companies is used as a separate risk factor to explain the returns on individual equities. In these models, small-cap equities are more risky than the equities of larger companies, which is counterbalanced by a higher expected return. A common explanation as to why equities of small companies would be more risky than other equities is that small companies will, on average, have a higher propensity for finding themselves in financial difficulties in times of weak economic development in society in general. Investors may therefore lose their capital during periods when they most need to generate a return, i.e. when the markets in general are weak. In order for investors to be willing to carry this risk, they have to be compensated by way of an additional risk premium. Norges Bank points out that this explanation is disputed in the academic literature. The Bank is of the view that the large, but varying, return differences between small and large companies over long periods of time suggest that the risk associated with small companies may constitute an independent and priced risk factor in the market. An investor like the Government Pension Fund, which is likely to have less of a need for liquidity than other investors during periods of weak economic development, should therefore, in the opinion of Norges Bank, be well placed to reap such a risk premium.

Source Norges Bank

The arguments Norges Bank believes may be invoked against the inclusion of small companies are of a more operational nature:

• The costs associated with purchasing the equity portfolio will increase. Norges Bank estimates the costs incurred through the initial purchases at about 0.06 pct. of the overall equity portfolio of the Fund if such purchases are spread over ten months, such as to reduce the market impact. Additional cost reductions may be achieved through using ongoing capital inflows for equity investments, or through coordinating the phase-in of a new benchmark portfolio with an increase in the equity portion. At the same time, a gradual change in the benchmark portfolio of the many sub-portfolios of the Fund make it challenging to extend the phase-in period over a very long period of time.

• The costs associated with the maintenance of the portfolio will also increase. This is partly because the costs incurred per transaction are higher for small-cap equities, but first and foremost because there are more frequent changes in the composition of the index. Updated estimates for this cost increase suggest that the costs relating to the maintenance of the benchmark portfolio for equities will increase by 0.03 – 0.04 percentage point. Norges Bank goes on to state that investors will normally require high transaction costs to be compensated by way of higher gross returns. How realistic this is in the market for small companies depends, as pointed out by Norges Bank, on how efficiently the market actually works.

• More than 4,500 new companies in the benchmark portfolio will also require more resources to be devoted to Norges Bank’s corporate governance efforts and to the Council on Ethics’ assessment work concerning the potential exclusion of individual companies pursuant to the Ethical Guidelines for the management of the Government Pension Fund – Global.

• The five pct. ownership limit may, according to Norges Bank, make it more difficult to increase the current holding of small-cap equities for purposes of maintaining the desired deviations from the benchmark portfolio.

The Strategy Council has addressed the issue of small companies in a letter of 20 February 2007 to the Ministry of Finance. The Strategy Council deems the proposed broadening of the benchmark portfolio to represent a natural extension of the current investment strategy. The Council is also of the view that broadening the benchmark portfolio for equities to include the small–cap segment is likely to be favourable from the perspective of the Fund’s return and risk, despite the higher costs implied by such a strategy. The Council also points out that a broadened benchmark portfolio will constitute a more appropriate comparative basis for the evaluation of Norges Bank’s active management performance in relation to the Fund.

3.3.3 The Ministry’s assessment

The objective of the investments of the Government Pension Fund – Global is to achieve the maximum possible return, given a moderate level of risk. To reduce the level of risk, the investments are spread across several asset classes, and each asset class contains several segments and sectors across many countries. At present, the benchmark portfolio for equities comprises in excess of 2,400 equities listed in market places in 27 countries. The value of the companies in the benchmark portfolio represents about 85 pct. of the overall value of the stock markets encompassed by the FTSE index. By including small-cap equities in the benchmark portfolio, as recommended by Norges Bank and the Strategy Council, one more segment will become included in the equity benchmark, with the result that the value of the companies included in the benchmark portfolio will represent about 96 pct. of the value of the stock markets encompassed by the FTSE index.

A practical approach to the assessment of such a broadening may be to split the life cycle of a business into the phases of growth and maturity. The growth companies are often found in the small-cap segment, whilst the largest businesses have entered a mature phase. It may be argued that a large and global fund like the Government Pension Fund – Global should be invested in broadly composed portfolio that encompasses both of these phases in the life cycle of companies.

A more theoretical perspective is that the world market portfolio, i.e. the portfolio of all equities in the world, offers the best trade-off between return and risk. The covariation between the return on small companies and that on large and medium-sized companies will be high, but they will not be fully aligned. This implies that a broadening of the benchmark portfolio will provide a somewhat more favourable ratio between average return and risk over time. This diversification benefit is expected to be moderate, but positive.

In addition to the diversification benefit as such, the recommendations from the Strategy Council and Norges Bank show that there is a possibility that the size of companies constitutes an independent and priced risk factor in the market. In such case the broadening will result in a further improvement to the Fund’s risk-adjusted return, because investors can expect compensation in the form of a higher expected return on small companies. However, it is uncertain whether such a risk factor exists.

The management of small-cap equities is more costly that managing the equities of larger companies. In its letter, the Strategy Council writes that this may explain why several other large funds have only included small-cap equities amongst their permitted investments, whilst refraining from including these in the benchmark portfolio. The costs associated with initial adaptation are estimated at about 0.06 pct. of the value of the equity portfolio if the adaptation is effected over ten months. A long time horizon is adopted when purchasing the equities, and the Ministry is of the view that these costs should not be perceived as an obstacle to broadening the benchmark portfolio.

However, annual management costs will also increase. This may be illustrated through the annual costs incurred in maintaining a portfolio equal to the benchmark portfolio, with and without the small-cap equity segment. Revised estimates from Norges Bank indicate that these maintenance costs will increase by 0.03-0.04 percentage point if small companies are included. This means, with an equity portion of 60 pct., that overall costs may increase by about 0.02 percentage point. In its assessment of this cost increase, the Ministry has attached weight to the argument that one would expect, even in the absence of any priced risk factor associated with the size of companies, the Fund’s risk-adjusted return to be maintained despite these increased costs.

Furthermore, the Strategy Council has pointed out, in its letter of 20 February 2007, that a broadened benchmark portfolio will constitute a more appropriate comparative basis for the evaluation of active management performance in relation to the Fund. The Strategy Council writes that:

“The Strategy Council believes that considerations to do with the evaluation of active management performance also favour the inclusion of small-cap equities in the benchmark portfolio. Since small companies form a large segment in the stock market, and since GPFG [Government Pension Fund – Global] already has access to such investments, a benchmark portfolio that includes small companies will constitute a more appropriate comparative basis than does the current benchmark portfolio.”

As part of its assessment of consequences in terms of the effort relating to the Ethical Guidelines of the Fund, the Ministry has requested the Council on Ethics to explain the implications for its work of the benchmark portfolio for equities being broadened through the inclusion of approximately 4,500 small companies. In a letter of 21 March 2007 to the Ministry of Finance, the Council on Ethics has highlighted three observations:

• The Council on Ethics writes that it will be possible to hire consultancy firms to monitor the portfolio through daily news searches and filtering the portfolio for companies that manufacture certain types of weapons, also when the portfolio is expanded to include small companies.

• The Council estimates that its annual costs will increase by about NOK 3 million because:

  • The price of monitoring the portfolio is expected to increase in proportion to the number of companies in the benchmark portfolio.

  • Less readily available information increases the need for hiring consultants with a stronger regional affiliation.

• The number of cases requiring additional assessment is expected to increase. It will therefore be necessary to strengthen the assessment capacity of the secretariat through, inter alia , the expanded use of external examiners to assist in the investigation of specific cases.

• Furthermore, the Council writes that: “When the number of companies increases, there is also an increased risk that the portfolio will include companies that are involved in activities that are in violation of the guidelines. Nor can the possibility that the availability of information is more restricted for small companies than for larger ones be ruled out. Nevertheless, the Council is not in possession of any information to suggest that investments in small companies as such are either more or less risky from an ethical perspective than other investments.”

Based on the Council on Ethics’ assessment of the consequences for its efforts relating to the deliberation of company exclusions, and Norges Bank’s assessment to the effect that the costs associated with corporate governance efforts in relation to a significantly larger number of companies will be limited relative to overall management costs, the Ministry is of the view that concern for the efforts relating to ethics should not be perceived as an obstacle to a broadening of the benchmark portfolio.

The Regulations relating to the management of the Government Pension Fund – Global currently operate with a 5 pct. upper limit on ownership stakes in individual companies. This limit has been fixed on the basis of, inter alia , an assessment of the consideration that a low ceiling on ownership stakes is a clear signal that the Pension Fund is a “financial”, as opposed to a “strategic”, investor. On the other hand, Norges Bank writes that an important strategy in seeking excess return is to identify businesses with a large profitability potential at an early stage in the development of such businesses, and these will typically be small and medium-sized enterprises. If small-cap equities are included in the benchmark portfolio, it is likely that the managers will wish to increase their holdings to maintain the size of the active positions. The 5 pct. limit on ownership stakes may make this more difficult.

It will, under any circumstance, be necessary to embark on a renewed assessment of the ceiling on ownership stakes in the not too distant future, due to the overall effect of the Fund’s strong growth and the increase in its equity portion. The Ministry is of the view that concern for the ceiling on ownership stakes should not prevent the benchmark portfolio from being broadened to include the small-cap segment. The Ministry will revert later with a renewed assessment of the upper limit on ownership stakes.

A final potential counterargument against the inclusion of small-cap equities in the benchmark portfolio at present is the market’s pricing of such equities. Traditional indicators relating to the pricing of companies, like the ratio between price and dividends, indicate that small companies have become more expensive in recent years, relative to large companies. In its analysis, Norges Bank points out the possibility that the price one has to pay for small companies may be unfavourably high. At the same, the Bank writes that these signals are too weak to make them relevant to the issue of whether small companies should be included in the benchmark portfolio. The Ministry agrees with this assessment. In addition, the risk of purchasing the equities at a point of time that turns out, in retrospect, to be unfavourable is reduced as a result of the benchmark portfolio changing over time, and as the result of the Fund’s investments in both large and small companies being spread out over time as the Fund received new inflows.

Against this background, the Ministry plans to broaden the benchmark portfolio of the Government Pension Fund – Global to include the small-cap equity segment of the FTSE index. The adjustment to the new benchmark portfolio will be implemented over a fairly long period of time.

3.4 The regulation of recognised market places and currencies for the Government Pension Fund – Global

3.4.1 Introduction

Section 4 of the Regulations relating to the management of the Government Pension Fund – Global stipulates a list of 42 permitted market places for equities (where the equities are listed) and the currencies of 31 permitted countries for bonds (which currency the bond is issued in). Within the list of permitted investments, the benchmark portfolio comprises indices for 27 stock markets and bond indices for the currencies of 21 countries.

The list of recognised markets and currencies for the Government Pension Fund – Global has been deliberated at regular intervals in the Budget documents, and was last expanded in 2003 in connection with the National Budget for 2004. The deliberation was based on input from Norges Bank.

The National Budget for 2006 discussed several changes to the framework governing Norges Bank’s management of the Fund. The new guidelines, which entered into effect on 1 January 2006, have less of an emphasis on the detailed regulation of which markets and instruments the Fund may invest in. In a letter of 11 March 2005, Norges Bank wrote, in the context of this rearrangement, that there was no need, from the perspective of the overall level of risk assumed by the Fund, for the Ministry to lay down detailed regulations as to which countries the Fund could be invested in either. At the same time, Norges Bank wrote that other considerations might nevertheless suggest that the Ministry should prepare a list of permitted market places and currencies. The National Budget for 2006 includes the following remarks in relation to Norges Bank’s proposal:

“The Ministry of Finance appreciates Norges Bank’s argument that the detailed regulation of which countries and currencies are permitted for investment is not required for purposes of managing overall risk. At the same time, other considerations favour the continued use of such a list. (…) The Ministry is of the view, based on an overall assessment, that the list of countries pertaining to the Petroleum Fund shall remain part of the guidelines. One intends to subject the list of countries to renewed examination, with a view to its potential expansion. One will in this effort be drawing on, inter alia , advice from Norges Bank. The Ministry will revert to this matter.”

In a letter of 13 July 2006, the Ministry requested input from Norges Bank for a renewed examination of the list of permitted market places and currencies. One also asked about Norges Bank’s views on whether it would be more appropriate to define the investment universe for fixed income through a list of recognised issuing countries instead of a list of currencies.

In a letter of 4 September 2006 to the Ministry, Norges Bank primarily recommends transition to an arrangement without any explicit list of permitted markets and currencies being set out in the Regulations, whereby it would instead be up to Norges Bank to prepare internal guidelines as to which markets and currencies the Fund may invest in. Alternatively, Norges Bank recommends that a new list be set out in the Regulations, expanded by 38 market places for equities and 49 currencies for fixed income. Consequently, Norges Bank recommends that the list, if any, should continue to define permitted countries for the listing of equities and currencies of permitted countries in which the bond portfolio may be invested. The letter from Norges Bank is available on the Ministry’s website (www.finansdepartementet.no).

3.4.2 The Ministry’s assessment

The Ministry is of the view that an expansion of the list of permitted markets and currencies for the Government Pension Fund – Global may increase Norges Bank’s opportunities for generating an excess return relative to the Fund’s benchmark portfolio. Moreover, the experience gained by Norges Bank through investments in several less developed markets may provide useful input for a future evaluation of whether the benchmark portfolio should be expanded to include new markets. At the same time, an increase in the number of countries should be combined with a clearer division of responsibility between Norges Bank and the Ministry of Finance as far as concerns assessments relating to where the Fund should be permitted to invest.

In its letter of 4 September 2006, Norges Bank proposes that the Bank should itself approve which markets the Fund may invest in. This recommendation is based on the reasoning that the economic assessment of markets and currencies may be delegated to the operational manager, within the general requirements concerning valuation, return measurement, as well as the management and control of risk, set out in the guidelines laid down by the Ministry. A significant increase in the number of permitted markets and currencies raises a number of issues:

• Both the supervision of financial activities and relevant legislation may be inferior to that in developed markets. The settlement and clearing systems in many of the countries may be of a lower quality, and both political and macroeconomic risk are of a different nature than in developed markets.

• Measurement of the Fund’s return requires the pricing of the securities in the portfolio. In new markets, the quality of pricing may be significantly lower than that in the markets where the Fund is already invested.

• Because risk measurements are in principle a matter of repeated pricing, one will be facing the same challenges when it comes to the measurement and management of market and credit risk as in the pricing or valuation of a portfolio. In addition, there has to exist information that enables the calculation of the price and risk associated with the security. For bonds this will, for example, be an overview of coupon payments, the loan repayment profile, as well as any options embedded in the bond. There will in many cases be no price history available for new markets. This will make it difficult both to measure risk and to price the instruments. For new markets it may, in addition, be more difficult to obtain supplementary information like repayment profile and options, etc.

The Ministry of Finance would therefore require Norges Bank to have completed, prior to the implementation of investments in new countries, a thorough process, which would in each individual case provide the bank with an overview of relevant issues relating to valuation, return measurement, as well as the management and control of risks associated with investments in each individual market and currency. Such effort would be required to include an assessment against the requirements implied by the Regulations and the supplementary guidelines, and to be documented. Provided that these requirements are met, the Ministry is of the view that there is no need for the detailed regulation of permitted markets on the part of the Ministry, as far as the management of the overall risk level of the Fund is concerned.

In the National Budget for 2006, the Ministry emphasised that other considerations favoured the continuation of the arrangement whereby a list of recognised markets and currencies was included in the guidelines of the Government Pension Fund – Global. In view of Norges Bank’s letter of 1 September 2006, the Ministry has undertaken a renewed assessment as to whether other considerations, like possible signal effects, ethical considerations or issues of a more political nature, suggest that the Ministry should be responsible for permitting investments in new markets and currencies. This review favours, in the view of the Ministry, Norges Bank’s primary recommendation, to the effect that Norges Bank should assume responsibility for preparing internal guidelines as to which markets and currencies the Fund may be invested in.

Ethical Guidelines for the Fund have been introduced since the last time the list was changed in 2003. The guidelines apply at company level, and cover activities in all countries where a company operates, independently of where the company is listed. The Graver Committee, which in the NOU 2003: 22 Green Paper laid the foundations for the current Ethical Guidelines, stated clearly that it would be difficult to justify, on the basis of ethical considerations, prohibitions on investments in the government bonds of certain countries, and that regular foreign policy channels represented a much more important tool for influencing the governments of other countries in the desired direction, than did exclusion from the Fund’s investment universe.

The list of permitted market places for equities and permitted currencies for fixed income is not suitable for purposes of excluding participation in activities in individual countries. A company listed in a recognised market place in a given country may, for example, have large parts of its activities concentrated in a country that is not included in the Fund’s investment universe. Furthermore, bonds issued by companies or states outside the list set out in the Regulations will be permitted as long as the bonds are issued in a currency that is included on the list of permitted currencies.

The Ministry proposes, on the basis of an overall assessment, that the current list of recognised markets and currencies be deleted from Section 4 of the Regulations relating to the management of the Government Pension Fund – Global. At the same time, Section 3.2 of the guidelines, on interest-bearing securities issued by the public sector in the currency of another country, will be amended, and Section 4 of the guidelines will include an explicit documentation requirement. This implies that Norges Bank will be required to document satisfactory routines for the assessment of permitted markets and currencies. The arrangement whereby the Council on Ethics evaluates the investments in equities and corporate bonds will be continued. Investments in equities and corporate bonds are subject to the Ethical Guidelines laid down by the Ministry, and the Council on Ethics will be responsible for rendering advice on the exclusion of companies in the new markets.

It is further intended to establish a procedure whereby the Ministry of Finance may prevent Norges Bank from investing in government bonds issued by specific countries. This will result in a clear division of responsibility between the Ministry and Norges Bank. Norges Bank will assess the financial aspects of the investments, whilst the list prepared by the Ministry establishes a procedure to ensure the exclusion of government bonds from individual countries if such exclusion enjoys broad political support. To avoid the creation of uncertainty as to the purpose of the Fund’s investments, it is emphasised that such decisions have to reflect broad political agreement in line with the principle of “overlapping consensus” as defined by the Graver Committee. Decisions not to invest in the government bonds of individual countries should therefore primarily apply to countries in respect of which UN sanctions have been imposed, or countries that are subject to other special international measures supported by Norway. The Government intends, against the background of the measures adopted by the EU and other countries against Burma (Myanmar), to amend the guidelines in such a way as to explicitly bar Norges Bank from investing the Pension Fund’s capital in bonds issued by the state of Burma. In formal terms, this will take place in connection with the amendment of the Regulations, cf. the discussion above.

3.5 Real estate and infrastructure in the Government Pension Fund – Global

3.5.1 Introduction

In the National Budget for 2007, it was pointed out that the largest asset class that currently falls outside the benchmark portfolio of the Government Pension Fund – Global is real estate, and that the Ministry will be embarking on an assessment of the scope for including this asset class in the benchmark portfolio. Most large institutional investors have invested parts of their portfolios in the property market. A report prepared for the Ministry by CEM Benchmarking shows that the world’s largest pension funds hold average investments of 6 pct in real estate. 4 These investments generate a return that depends on the price developments of, and rent incomes from, office buildings, shopping centres, industrial buildings and, to a certain extent, other types of property as well. A low degree of covariation between the return on real estate and the return on equities and fixed income, a high direct return in the form of rent income, and a desire to protect the portfolio against inflation, are often highlighted as motivating other funds’ investments in real estate.

Although the Pension Fund – Global currently has no separate allocation to real estate, the Fund has exposure to the property market through its investments in listed real estate equities, which are included in the Fund’s benchmark portfolio. Just below 2 pct. of the market value of the Fund’s benchmark portfolio for equities is comprised of real estate equities. In addition, many of the companies in the Fund’s equity portfolio will own their own production and office premises. It is difficult to determine the impact of this on the risk profile of the equity portfolio.

The size of the market for listed real estate equities is on the increase, not least because an ever-increasing number of countries are facilitating the establishment of tax-transparent investment companies in the property market, so-called REITs 5 , which are aimed at institutional and private investors. However, the unlisted market is much larger than the listed one, and is likely to remain so for a long time to come. A significant portion of the real estate portfolios of other large funds is therefore held in the form of unlisted investments. 6 Such investments are often illiquid. It is, for example, not uncommon for investments in unlisted real estate funds to in practise imply that one binds the invested capital for more than ten years. One may sell one’s interests in the fund in the secondary market, but will then have to expect to sell such interests at a not insignificant discount from their real value. It is commonly assumed that investors are compensated, in the form of a higher expected return, when investing in less marketable assets. This excess return is often labelled an illiquidity premium. The Government Pension Fund adopts a long investment horizon, and the probability that it will have to sell its holdings on short notice is low. Generally speaking, the Fund is well placed to reap such illiquidity premiums.

Investments in infrastructure projects, like electricity and water supply, toll roads, airports and telecommunications, have traditionally represented a limited market. However, increasing participation and funding from the private sector in recent years have made such investments attractive for long-term financial investors. Although this market is expanding, and is expected to undergo considerable growth in coming years, it remains small compared to the property market. Nevertheless, it would be appropriate to examine such investments concurrently with an examination of investments in real estate, since the return and risk characteristics of infrastructure investments as similar to those of traditional real estate.

3.5.2 Norges Bank’s recommendation

In its letter of 20 October 2006 to the Ministry of Finance, Norges Bank has, inter alia , recommended that the Government Pension Fund – Global be invested in real estate and infrastructure.

Norges Bank is of the view that there are two main arguments in favour of including real estate and infrastructure in the Fund’s benchmark portfolio. Firstly, the Pension Fund is well placed to reap illiquidity premiums. Secondly, investments in real estate and infrastructure imply that the risks assumed by the Fund are spread across more asset classes. This implies that such investments may reduce the level of risk assumed by the Fund, without reducing its expected return.

It is commonly assumed that investments in real estate are characterised by average returns and risks that in the long run fall between the levels associated with equities and bonds. The reason why such investments may nevertheless reduce the overall risk level of the Fund, without reducing its expected return, is that the return on real estate is influenced by factors that are peculiar to that market, or that factors which influence the return on other asset classes have a different impact on the property market. This implies that the returns on investments in the property markets and in the other capital markets will not be fully correlated.

Another important reason why funds tend to hold investments in real estate is that such investments are expected to generate a stable real return over time, i.e. to offer good protection against surprising changes in inflation. For funds carrying explicit liabilities that will increase in line with inflation, the combination of inflation hedging and higher returns than on inflation-hedged bonds will be valuable. According to Norges Bank, this is of lesser importance to the Government Pension Fund. Besides, the degree of inflation hedging will vary between markets and projects, and particular demand and supply conditions in the property markets may also limit such effect. Norges Bank is therefore of the view that inflation hedging considerations should not be accorded much weight in assessing a separate allocation to real estate.

The management costs incurred in relation to investments in the property market will be significantly higher than those associated with management in listed markets. CEM’s cost report for 2005 shows that the funds comprising the group of comparable funds incurred, on average, annual management costs of just below one percentage point in relation to unlisted, externally managed real estate.

Norges Bank writes that it will probably not be practicable, within the foreseeable future, to achieve a higher exposure to real estate and infrastructure than one corresponding to about 10 pct. of the overall portfolio. The model computations of the Bank confirm that such an allocation may result in a better diversification of the Fund’s risk. If such investment is matched by correspondingly lower investments in fixed income, the overall risk level of the Government Pension Fund – Global would be more or less unchanged, whilst the expected return on the portfolio would increase. Norges Bank further writes that:

“Norges Bank recommends that, in the light of the probable liquidity premium and the probable diversification gains from investing in real estate and infrastructure, it be made a long-term strategic target for up to 10 per cent of the Government Pension Fund – Global to be invested in this asset class. This portfolio will have to be built up gradually over a period of several years.”

Finally, the letter points out that it would be necessary to define an investment mandate that specifies the required rate of return, risk limits and reporting requirements before embarking on any investments in real estate and infrastructure.

3.5.3 The Ministry’s assessment

The objective of the investments of the Government Pension Fund is to achieve the maximum possible return, at a moderate level of risk, cf. above. The scope for further diversifying the risk associated with the investments by including investments in real estate and infrastructure, and the scope for reaping a gain over time by investing in less marketable investments, should therefore be examined more closely.

The Ministry has requested the Strategy Council to evaluate whether the return and risk characteristics associated with investments in real estate suggest that such investments should form part of the benchmark portfolio of the Fund. The Strategy Council has commissioned external real estate management experts to prepare a report on investments in real estate. This report will, together with the analysis performed by Norges Bank, provide valuable inputs to the evaluation to be carried out by the Strategy Council.

Before it can be decided whether real estate and infrastructure should be covered by the Fund, it is also necessary, in addition to the technical investment considerations, to shed more light on the challenges in terms of administrative control. The manner in which the equity and bond portfolios of the Government Pension Fund – Global are managed at present, with a benchmark portfolio and an upper limit on tracking error, will not be directly applicable to a portfolio comprising real estate and infrastructure investments, irrespective of how such management is organised. The fixed-income and equity indices used by the Pension Fund are updated on a daily basis. Those indices that are available in the property market are updated monthly, quarterly or annually. Values will mainly be based on appraisals, and only to a lesser extent on completed transactions. The index values will therefore be uncertain, and difficult to use as a standard of comparison for the actual performance of this part of the portfolio. Nor will the indices be good management tools for actual investments because, inter alia , in many countries these only reflect developments in a limited part of the property market.

International investments in real estate will involve the manager being granted wider authority than is the case for Norges Bank’s management of the current equity and bond investments. Such an assignment would imply less control over the contents of the real estate portfolio on the part of the Ministry, which would instead have to place more of an emphasis on reporting requirements in combination with management evaluations in cooperation with external experts.

Since a large share of the return is in the form of direct returns from rent income, the assessment of tax implications is an important aspect of real estate management on the part of other funds, and may impact on the composition of the portfolio. The Ministry will therefore have to examine the tax implications of real estate investments in the Government Pension Fund – Global in more detail.

In order to ensure that all aspects of this issue are illuminated, the Ministry will initiate several studies relating to expected return and risk, the organisation of the management effort, as well as the management model. Based on such work, the Ministry will revert with an assessment as to whether real estate and infrastructure should be included amongst the Fund’s investments.