Meld. St. 21 (2014–2015)

The Management of the Government Pension Fund in 2014 — Meld. St. 21 (2014–2015) Report to the Storting (white paper)

To table of content

3 Unlisted investments in the GPFN

Letter of 13 November 2014 from Folketrygdfondet to the Ministry of Finance

Summary

We refer to the letter of 27 June 2014 from the Ministry of Finance, in which Folketrygdfondet is requested to analyse and assess whether unlisted investments such as real estate and infrastructure should be included in the management of the Government Pension Fund Norway (GPFN).

Folketrygdfondet is of the view, based on an overall assessment, that permitting unlisted real estate and infrastructure investments may be positive for the GPFN. These are illiquid assets that are, generally speaking, well suited for a long-term fund with limited liquidity needs and a competent asset manager. We believe, based on the distinctive characteristics and expertise of Folketrygdfondet, that such an expansion of the investment universe may serve to diversify risk in the Fund and increase its return net of costs.

Unlisted assets differ from the more liquid investment alternatives currently open to the GPFN in that they may be difficult to divest, involve significant transaction costs and present uncertainties in terms of ongoing valuations. Existing data make it challenging to conclude unequivocally that the inclusion of unlisted assets will generally deliver a better risk-adjusted return on a portfolio, but we are of the view that this asset class will, over time, include investments that can serve to improve the ratio between risk and return in the GPFN. However, in order for investments in unlisted assets to strengthen the risk-adjusted return on the GPFN over time, it is our view that it will be necessary to have a governance structure and a mandate that facilitate investment decisions based on a long-term perspective and with a sharp focus on costs and the scope for safeguarding our share of investment value. We believe, in view of the special characteristics of unlisted asset investment in our investment universe, that such considerations are best attended to by extensive delegation of investment decisions to the asset manager, i.e. that unlisted asset investment takes place as part of our active management, rather than by the Ministry defining a strategic allocation for unlisted investments in the benchmark index.

We envisage that Folketrygdfondet will be a selective investor, collaborating with other market participants with special developmental and operational expertise when making unlisted asset investments, with the development of an unlisted asset portfolio taking place gradually and over time as we accumulate expertise and experience and identify suitable investments. We are of the view that unlisted asset investment can be handled within the current general risk limit as expressed in terms of expected tracking error, and recommend that unlisted assets as a portion of the GPFN be capped at 10 percent.

An expansion of the investment universe to include unlisted real estate and infrastructure investments will increase the scope for manoeuvre in Folketrygdfondet’s long-term active management. Unlisted asset investment will result in a moderate increase in the costs of Folketrygdfondet, but we believe that the increase in excess return will be larger. Unlisted investments will, at the same time, serve to diversify active management risk.

The current organisation has relevant expertise on general asset management, listed equity investments within real estate and fixed-income investment within real estate and infrastructure. Folketrygdfondet has extensive experience from investing a large fund in a financial market with, at times, weak liquidity. This has given Folketrygdfondet a long-term asset management perspective, which is necessary for the management of unlisted investments. There will, in addition, be a need for strengthening our expertise in handling the special challenges involved in deciding and following up on relatively large, tailor-made transactions in illiquid assets. We are of the view that such activities should be integrated into the current organisation, sharing its organisational culture and values.

Whilst the market for unlisted real estate investments is relatively well-developed and has a long investment history, the market for infrastructure investments remains immature. If it is desirable to expand the infrastructure market in Norway, such expansion will, inter alia, depend on decisions made by various government bodies concerning the scope of infrastructure investments and public ownership.

Introduction

The purpose of this letter is to provide input to enable the Ministry to decide whether the mandate for the GPFN should, in principle, be expanded to include unlisted investments like real estate and infrastructure. If the Ministry decides that such expansion is merited, important preparations remain before the first unlisted investments can, if applicable, be made. We will in such a process comment more comprehensively on detailed matters raised by the Ministry in its letter, including investment strategy for unlisted assets, risk management, organisation, costs, operational implementation, follow-up and reporting.

We have in our analyses and assessments focused on the market for unlisted investments in Norway. In developing new investment activities we are committed to a gradual approach and the accumulation of experience. Consequently, we deem it appropriate to initially focus on Norway, where we have the most expertise and experience, and consider expansion into the other Nordic countries in our investment universe at a later stage.1 Moreover, we focus, in line with the Ministry’s letter, on unlisted real estate and infrastructure investments (and not on unlisted equity investments in the form of private equity, seed capital, etc.). In addition, we have devoted special attention to the specific governance challenges posed by unlisted and illiquid investments, since the available data offer less scope for drawing clear conclusions in relation to the general return, risk and correlation characteristics of this asset class than in relation to listed investments.

We have, in examining this matter, consulted investors that we find it reasonable to compare ourselves with, as well as independent experts. We have commissioned two external reports on real estate and infrastructure investments from Akershus Eiendom and Pöyry, respectively. These are enclosed with this letter.

It follows from the external reports that some documented experience is available for the purposes of examining the Norwegian market for commercial property from a professional investment perspective. The market for infrastructure is, on the other hand, immature and fragmented, although ongoing processes may potentially result in more developed markets over time. However, analyses based on international experience are available, and these may in a globalised world be assumed to be of relevance in a Norwegian context as well. Reference is made, in particular, to the thorough assessment by the Ministry of Finance of real estate investments and illiquidity premiums in the Government Pension Fund Global, for example in Reports No. 16 (2007–2008) and No. 27 (2012–13) to the Storting, as well as the discussion note on infrastructure prepared by Norges Bank Investment Management (NBIM discussion note #2–2013). A report prepared for the Ministry by CEM Benchmarking also provides some information on asset allocations, returns and costs for comparable funds.2

Unlisted investments in general

Investments in unlisted real estate and infrastructure are generally characterised as being illiquid and having relatively high transaction costs, as well as there existing uncertainty as to their ongoing valuation. Such unlisted investments may be very difficult to divest from in the short-term, but will in most cases give rise to a relatively stable and secure cash flow over several years. The investments tend to be heterogeneous with different risk characteristics, which reflect both the uncertainties in the individual projects (for example in relation to the tender phase, construction phase, ongoing management, the willingness of lessees to pay, equity portion, funding model) and uncertainties of a more general nature (for example politics and regulations, macroeconomic developments, risk-free interest rate, risk aversion). It can be challenging to establish clear definitions as to whether a project should be categorised as real estate or infrastructure. The processes leading up to an investment are typically labour-intensive and time-consuming, requiring local knowledge and extensive use of legal expertise. The transactions can be relatively large, involve special corporate structures and shareholders’ agreements, and have a funding structure involving different equity and debt tranches. Considerable costs can be incurred in preparing an investment, whilst the return is spread over a long period of time.

Unlisted real estate and infrastructure investments comprise a relatively common asset class amongst long-term investors internationally. CEM’s report shows a total allocation of close to 10 percent on average for funds with which it is reasonable to compare the GPFN, whilst the allocation can be considerably higher for funds with a strong belief in this asset class. Real estate and infrastructure are deemed by market participants to offer return and risk characteristics between those of bonds and equities, with part of the return being attributable to the compensation required by investors for weak liquidity (illiquidity premium). Moreover, several market participants emphasise that this is accompanied by a specific type of risk exposure that cannot be achieved through a simple combination of listed equities and bonds, thus implying that the inclusion of unlisted assets can serve to improve the ratio between return and risk in a portfolio. It is, however, challenging to draw clear conclusions concerning the return and risk characteristics of unlisted assets in general on the basis of statistical analyses.3

In addition, some investors state that they recognise and value that the asset class offers better inflation protection than nominal bonds. Besides, unlisted real estate and infrastructure investments provide exposure to stable cash flows, whilst ongoing valuations are less variable than those of liquid instruments, and these characteristics are valued by some investors. Favourable tax treatment may also motivate certain investors to invest in unlisted assets.

There are different methods for investing in real estate and infrastructure, and the various alternatives involve exposure to different risk factors and entail different asset management costs. The table illustrates this as a choice between listed and unlisted investments and between direct and indirect investments. Unlisted investments will typically involve higher liquidity risk and asset management costs than listed investments in such a matrix, whilst direct investments will entail higher project-specific and regulatory risk than indirect investments.

Table 3.1 Different investment alternatives

Direct

Indirect

Unlisted

Direct investments, project finance

Unlisted funds

Listed

Equities, bonds

Listed funds

The types of investments chosen by an investor will depend on which alternatives are available, as well as on the special characteristics and preferences of the investor. In Norway, there has generally been little availability of listed equities offering direct exposure to real estate and infrastructure.

In assessing whether unlisted real estate and infrastructure investments are suitable for the GPFN, Folketrygdfondet will consider and attach the most value to the contribution from this asset class towards improving the ratio between risk and expected return (net of costs) for the Fund. We therefore give little emphasis to considerations like inflation protection, more stable return measurement and tax considerations.4 Moreover, we have noted international experience that highlights the importance of achieving the exposure without incurring excessive costs,5 i.e. that caution should generally be exercised regarding investments in structures that involve many levels and intermediaries.

Some data are available for returns on Norwegian commercial properties over time, as well as some experience from infrastructure projects, but a robust assessment of the returns, risk and correlation characteristics of real estate and infrastructure will to a large extent have to be based on international experience. Internationally, the data available in the public domain is also limited, and few studies have been published on the return and risk characteristics of unlisted assets, especially for infrastructure. Besides, historical returns on unlisted asset indices tend to underestimate risk. When compared to listed asset indices, unlisted asset indices are backed by few trades and valuations often utilise appraisals or methods that result in minor variations from one period to the next. Such indices are often based on a small number of unleveraged investments, which will not always be representative of the return characteristics of the actual projects invested in by an investor. Correspondingly, one should be cautious about expecting diversification gains as large as may be suggested by a comparison of return series for unlisted assets with those for listed equities and bonds.

In summary, we assume that the expected return and risk of unlisted assets are generally somewhere between those of listed equities and bonds, although there are large variations internally within the asset class that can be linked to project-specific considerations. Our perspective is that this asset class will, over time, offer investment opportunities that can serve to increase the risk-adjusted return on the portfolio.

The market for unlisted real estate investments

Commercial property is a general term for all property not used as one’s own home. This is a heterogeneous sector comprising a mixed set of investment opportunities involving varying risk exposure and a variety of market participants. Key segments include office premises, retail, hotels, warehousing and logistics, manufacturing and production premises, health and schools, residential letting, as well as facilities that are bordering on infrastructure. The largest segments for investors are office premises and retail. Residential leasing is perceived as a category of interest to financial investors internationally, since this segment offers a different risk exposure than other commercial property. However, the availability of such properties in Norway has been limited for financial investors.6 There is a broad range of investor categories, and it has also been noted that tax considerations may be a motivating factor for some investors.

One alternative for investors is to invest in listed companies offering real estate exposure. There is, however, not much of a tradition for listed real estate companies in Norway. The only such companies currently listed on the Oslo Stock Exchange are Norwegian Property, Olav Thon Eiendomsselskap and Entra Eiendom. Norwegian Property holds properties valued at approximately NOK 14 billion, whilst Entra is almost twice as large and the leading office property investor in Oslo, with a market share of close to 6 percent. Olav Thon Eiendomsselskap has shopping centres as its main exposure. Norwegian Property and Olav Thon have for quite some time been traded at relatively large discounts on underlying asset values. It is likely that this is partly caused by weak liquidity in these equities, but other company-specific factors have also been mentioned (such as unfavourable transactions, aspects of the implementation of investment projects, vulnerable funding structures, etc.). It is likely that investors have also noticed that other listed real estate companies (Linstow, Avantor) have been delisted at prices that are lower than underlying asset values.

A key issue for a large investor like Folketrygdfondet is whether we can achieve significant real estate exposure for the capital of the GPFN within a reasonable period of time if we so choose. Based on the fact that sales of large, unlisted commercial properties in Norway over the last decade have averaged just under NOK 40 billion per year, it would appear to be feasible to establish a significant real estate exposure within a few years if this is deemed to be appropriate.7

The pricing of unlisted assets is influenced by factors that are also important for the listed stock market and the bond market. Reductions in the risk-free interest rate and risk premiums in recent years have lifted the value of stock markets and bond markets, as well as unlisted assets. This means, at the same time, that historical return series are likely to overestimate the expected return on both unlisted and listed assets in coming years. An analysis from Swedbank8 reports that office properties in Oslo have delivered an average annual return of 11 percent since 1983, which is 3 percentage points less than equities and 2 percentage points more than bonds. Furthermore, the analysis estimates an expected future return of 6 percent (assuming fixed real rent levels and no further re-pricing of the asset class), which is somewhat less than for equities and somewhat more than for bonds.

The market for unlisted infrastructure investments

Infrastructure investments include networks of fixed installations that supply services of key importance to the functioning of society (for example roads, railways, power grids, water and sewage), thus implying that the frequency of use is not particularly sensitive to cyclical fluctuations. Capital costs are often high, and suppliers operate in virtual monopolies in which revenues are subject to government regulation. Some countries have quite extensive experience with private infrastructure ownership (for example Australia, the UK and the US), whilst infrastructure has been associated with public ownership in most other countries. However, a need for consolidation of government finances has, in combination with a large and unmet investment need, resulted in expanded scope for private infrastructure ownership. Some observers also point to potential efficiency gains from involving private interests, whilst others highlight related governance challenges for the public sector.

Many asset managers with a long time horizon consider infrastructure investments to be of interest because of characteristics like stable cash flows over long periods of time, providing an element of inflation protection, and security in physical assets. Norwegian infrastructure is largely in public ownership, and Norwegian institutional investors hold few infrastructure investments in Norway.

There is reason to assume, at the general level, that a number of the factors that favour investing in real estate also pertain to infrastructure, although infrastructure projects also have certain special characteristics. It follows from the NBIM discussion note on infrastructure that the preparation of representative return series for infrastructure projects involves considerable challenges. The return series prepared internationally generally show that investment in infrastructure assets has improved the risk and return characteristics of a portfolio, but variations between the various projects are large and one should be cautious about basing future projections on overly optimistic assumptions.

The market for infrastructure investments in Norway remains immature. If it is desirable to develop the market for unlisted infrastructure investments, Folketrygdfondet will be able to contribute in a manner similar to our approach in listed markets. We promote well-functioning and efficient market places, as well as good standards and appropriate conduct from issuers, managers and investors, in line with the responsible investment provision in the mandate for the GPFN.

It is, however, important to emphasise that the key to developing the market for infrastructure lies with various government bodies: the supply of infrastructure assets depends on decisions concerning the scope of infrastructure investments and public ownership, and the demand for infrastructure assets from long-term financial investors depends on the government regulatory framework the investors operate within, as well as their assessment of regulatory and political risk.

Infrastructure investments differ from real estate investments inasmuch as revenues are more influenced by government-stipulated rates and regulations. Since these are typically projects of a long duration, it means that investors’ assessment of the degree of predictability in the exercise of governmental powers will be a key factor behind their willingness to provide capital and the pricing of the assets.9 The Pöyry report notes that there are currently some obstacles that need to be addressed before there is a good basis for developing a Norwegian infrastructure market with significant participation from financial investors. According to the report, international pension and infrastructure funds, whose interest and participation are desirable if one would like to develop a market place in line with good international practice, are somewhat cautious in their assessment of Norwegian infrastructure projects. Moreover, Norwegian life insurance companies and pension funds are governed by the Insurance Activities Act and the appurtenant Asset Management Regulations, which impose clear limitations on the scope of the infrastructure investments of such investors.

The distinctive characteristics of Folketrygdfondet

In Folketrygdfondets strategic plan for the management of the GPFN,10 we emphasize that our distinctive characteristics compared to other investors are of relevance to how we should focus our investment activities. Correspondingly, our assessment of unlisted investments takes place from the perspective of such characteristics and potential advantages. The GPFN and the manager, Folketrygdfondet, have a government owner and a long time horizon. We have a clear and sound framework that facilitates asset management in appropriate pursuance of the Fund’s objective, and we are often perceived as an attractive collaboration partner by other market participants. Management of society’s assets means, at the same time, that expectations as to efficient and responsible operations are high, and that we need to pay considerable attention to the importance of maintaining a good reputation and trust, not only on the part of our owner, but also in the market and with the general public.

The asset management framework suggests an investment behaviour characterised by a longer time horizon than many other investors. We can premise our management of the GPFN on a lesser need for liquidity, a higher capacity for absorbing risk and more tolerance for large fluctuations in value. This facilitates long-term countercyclical behaviour, which is a challenging, but profitable, investment approach that requires a unified and robust strategy with solid support both internally within Folketrygdfondet and externally from the owner and the general public. At the same time, a long time horizon means that we are well positioned to invest in financial markets that are at times illiquid, thus reaping a specific illiquidity premium. The magnitude of such illiquidity premiums varies over time, implying that we need to engage in patient, thorough and active management and be prepared for the portfolio to deviate somewhat from the benchmark index for extended periods of time.

Folketrygdfondet invests the GPFN in its own name, and is amongst the largest individual institutional investors in the Norwegian financial market. Assets under management represent close to 5 percent of the listed stock market (10 percent when adjusting for free float) and almost 3 percent of the fixed-income market. This means that it takes a long time to effect large changes in portfolio exposure, and that we have a special interest in adopting a long-term perspective in our investment activities. Size also enables us to participate in large transactions and to exploit economies of scale in asset management, whilst giving us some scope for influencing the investment framework and functions offered by market places. A large, long-term investor like Folketrygdfondet is committed to responsible investment. Active and responsible exercise of our ownership and creditor rights is important for attending to our financial interests. In addition, our responsibile investment activities will help promote a more well-functioning market place and reduce environmental risk and social risk in our investment universe.

Folketrygdfondet has in the course of its 20 years of equity management and 40 years of fixed-income management developed broad asset management expertise and accumulated extensive experience from the Norwegian capital market. In assessing a potential expansion into unlisted real estate and infrastructure investments, we emphasise the fact that we are already permitted to invest in unlisted equities of companies that intend to obtain a listing, and also that we have experience from investment in listed real estate companies and bonds relating to real estate and infrastructure investments. Folketrygdfondet has extensive experience from investing a large fund in a financial market with, at times, weak liquidity. This has given Folketrygdfondet a long-term perspective on asset management, which is necessary in the management of unlisted investments.

In addition, there will be a need for strengthening our expertise in the handling of large, tailor-made transactions involving significant direct ownership stakes in markets where local knowledge is a distinct advantage. In such an asset class, which is characterised by special company structures tailored to the nature of the projects and where co-investors and collaboration partners may have different expectations and interests, we will emphasise the necessity of having good shareholders’ agreements that protect our interests and safeguard our share of cash flows and any achieved increases in value.

If we should attempt to idenity potential advantages that Folketrygdfondet would have as a real estate investor, we belive that our size, our high capacity to absorb risk and our responsible investment profile may make us an attractive collaborator for market participants that want a serious long-term co-investor in large projects. We respect, at the same time, that this is a fairly well-developed market with established market participants that have an informational advantage over a new investor.

In addition to drawing on the expertise and distinctive characteristics of Folketrygdfondet compared to other participants in the real estate and infrastructure markets, we expect to add value through allocation decisions. Given our long-term and countercyclical investment strategy, it will be positive to have access to an asset class offering investment opportunities with risk characteristics that differ from those of the investment alternatives available in our current investment universe.

Qualities such as size, a high capacity to absorb risk and a responsible investment profile can make us an attractive collaborator for market participants that want a serious long-term investor for large projects in the infrastructure market as well. Whilst there are international asset managers with more targeted expertise within special infrastructure segments, our experience from Norway can potentially represent an advantage that may be appreciated by both potential providers of infrastructure projects and co-investors. Our extensive experience from fixed-income investments, including in infrastructure projects, means that we will be in a position to contribute to the development of a more well-functioning infrastructure market.

Our characteristic of being a government-owned asset manager may also pose certain specific challenges in a marketplace where government bodies play a key role. Although Folketrygdfondet’s investment behaviour will be exclusively motivated by financial return and risk assessments in line with our mandate, our government ownership may give rise to expectations that we will also take other considerations into account. One may also envisage elements of reputational risk for Folketrygdfondet if we were to be invested in an infrastructure project of considerable interest to the general public that may also attract negative attention. It should also be noted that our assessment as to whether the return and risk characteristics of a project are attractive will often need to be mirrored by a willingness to pay on the part of a governmental project owner. If Folketrygdfondet is going to invest parts of the GPFN in infrastructure, it is important for the premises underpinning such activities to be firmly entrenched.

Organisation and governance

Folketrygdfondet’s recommendation that the GPFN may also be invested in unlisted real estate and infrastructure is conditional upon the establishment of an appropriate mandate reflecting the special characteristics of this asset class. Unlisted assets tend to be difficult to divest, incur high transaction costs and involve more uncertainty in relation to valuation, which means that investors expect an additional compensation (illiquidity premium) compared to investment in more liquid listed markets. In order for unlisted asset investment to strengthen the risk-adjusted return on the GPFN over time, it is necessary for the investment decisions to be made from a long-term perspective and with a sharp focus on the investment costs and the scope for safeguarding our share of the investment value. We are of the view that successful management of unlisted assets requires a high degree of delegation to the asset manager, good general control parametersand trust between the capital owner and the asset manager.

It is our recommendation that the exposure of the GPFN to unlisted assets take the form of Folketrygdfondet being permitted to invest in such assets whilst keeping the current benchmark index (comprising 60 percent equities and 40 percent bonds) unchanged, i.e. that the Ministry should not stipulate a strategic allocation for unlisted investments in the benchmark index. Such a governance model stimulates long-term management with an emphasis on achieving the maximum possible return net of costs, at a moderate risk for the GPFN as a whole, whilst at the same time acknowledging that there is uncertainty about the size of the market for unlisted investments in coming years.

We emphasise that an expansion of the investment universe to include unlisted assets will expand the scope of active management into areas that are well suited for a large long-term investor like Folketrygdfondet. We believe that our distinctive characteristics and advantages, our values, our investment philosophy and our extensive experience provide us, all in all, with a solid foundation for the continued generation of excess returns at low operating costs.

Folketrygdfondet’s active management has from 1998 until the end of the 3rd quarter of 2014 contributed an average annual excess return of 0.49 percentage points. Our active risk taking (deviations from the benchmark index) has, for various reasons, declined in recent years to a level that is unlikely to be consistent with equally high excess returns in coming years, but we assume that new investment opportunities that can give rise to more risk taking will materialise over time.11 If our investment universe is expanded to include unlisted real estate and infrastructure, it is reasonable to expect that this will, over time, serve to increase our expected excess return (net of costs) and add to the diversification of risk in active management. A more long-term focus in active management will, at the same time, entail longer periods of deviations between returns on the GPFN and on the benchmark index, and it is a prerequisite for success that this is acknowledged by both the capital owner and the asset manager.

We operate on the assumption that the current general risk limit, as defined by expected tracking error, will remain a suitable limitation over the next few years as well, including if unlisted asset investment is permitted. Since this asset class is characterised by weak liquidity and valuation challenges, operational risk management will, in comparison with listed instruments, also have to rely more on other risk measures than expected tracking error. The impact of unlisted asset investment in terms of the risk measure ‘expected tracking error’ may be perceived as greater than would be suggested by a more general risk assessment.12 The asset class may include both equity investments and fixed-income investments.

We recommend that the mandate stipulate an upper limit on the portion of unlisted assets in the GPFN, which will serve to further restrict the scale of such investments. It is our perspective that such a cap should offer sufficient scope for manoeuvre to enable the benefits desired from such asset management to be reaped, and we recommend that said cap be put at 10 percent. It is assumed that the establishment and development of Folketrygdfondet’s investments in this asset class will take place over an extended period of time, in line with our accumulating expertise, gaining experience and identifying suitable investments. Which types of investment projects we will focus on will, in the same manner, evolve over time. Our perspective is that we will be a selective investor and invest in collaboration with other market participants that have appropriate developmental and operational expertise. We will emphasise the identification of cost-effective investment models, in line with our investment philosophy.

Successful management of unlisted assets will require a moderate increase in operating costs. However, Folketrygdfondet aims to continue to be characterised as a cost-effective asset manager. We anticipate a moderate increase in staffing to develop core expertise within unlisted investments, which will then be able to draw on external resources when needed. We respect that core expertise within unlisted investments may come from an investment culture that differs somewhat from the current asset management organisation, which is focused on liquid and listed markets, but we expect the management of unlisted assets to be integrated into the current organisation, with a shared organisational culture and values.13

Unlisted investments will only be made if investments are available that, in the view of Folketrygdfondet, serve to improve the overall composition of the portfolio and that are attractively priced relative to other comparable investment alternatives available to the GPFN. Consequently, an opportunity cost model14 appears to be a flexible and useful tool in Folketrygdfondet’s asset management, for the purpose of facilitating disciplined long-term investments in unlisted assets. Such a model involves the establishment of a return measure for an unlisted investment project on the basis of the risk characteristics of the project in relation to listed equity and fixed-income indices. This return measure expresses the opportunity cost of the unlisted project, and is at the same time a tool for determining the extent to which one should fund an unlisted project by reducing equity holdings or bond holdings. However, good control over investment decisions in unlisted assets with a long investment horizon is challenging, and various control models involve their own advantages and disadvantages. The opportunity cost model is intended to provide incentives that are suited to promoting good long-term investment decisions. It may nonetheless be challenging to assess the risk and return on unlisted investments in the short term, since their value does not vary in line with the return measure (as the latter is linked to liquid markets in which valuations are more volatile in the short term than valuations of unlisted projects).

Our perspective is that unlisted asset investment shall contribute to the diversification of risk for the GPFN as a whole, and we recommend that no specific diversification requirement be applied internally within the unlisted asset class.

In order to ensure confidence in this particular element of asset management activities, we will emphasise the importance of disclosing and communicating what type of investment risk is involved, alongside the reporting of holdings, returns and suitable risk measures. The special characteristics of unlisted investments suggest that these should be managed and reported as a distinct asset class, separately from the current equity portfolio and fixed-income portfolio. We will, furthermore, proceed on the basis that the relevant principles for valuation, return measurement and risk management shall, at a minimum, be in conformity with internationally recognised standards and methods. The responsible investment principles of Folketrygdfondet will apply to this asset class as well, and asset management will reflect the special characteristics of the asset class. Our investor role in unlisted assets will, in certain respects, differ somewhat from our role in listed markets. Large ownership stakes suggest that we need to consider closer and more direct exercise of ownership, including through directorships, in order to attend to our financial interests.

If the Ministry decides that the mandate of the GPFN shall be expanded to include unlisted real estate and infrastructure investments, important preparations remain to be made by Folketrygdfondet before the first unlisted investments can be initiated. Key preparations include the establishment of specialist expertise and the preparation of a strategic plan for unlisted investments. We will in such a process comment more comprehensively on the investment strategy for unlisted assets (including the trade-off between investing in development projects and turnkey projects), risk management, organisation, costs, operational implementation, follow-up and reporting.

It will also be necessary to consider certain modifications to other asset management activities when investing in unlisted assets, to ensure that the overall exposure of the GPFN (for example to illiquidity) is in line with our objective.

Yours faithfully

Folketrygdfondet

Erik Keiserud

Olaug Svarva

Chairperson of the Board of Directors

Chief Executive Officer

Enclosures

Footnotes

1.

See also the current benchmark index for the GPFN, which has allocations of 85 percent in Norway and 15 percent in Denmark, Finland and Sweden.

2.

The sources drawn on in this letter are, unless otherwise stated, the reports from Akershus Eiendom, Pöyry and CEM, as well as Reports No. 16 (2007–2008) and No. 27 (2012–2013) to the Storting and NBIM discussion note #2–2013.

3.

In Report No. 27 (2012–2013) to the Storting, the Ministry of Finance refers, for example, to a report examining the potential for reaping illiquidity premiums in different asset classes, in which the authors note that the available data are not good enough to determine whether illiquidity premiums exist in unlisted markets like real estate.

4.

If such considerations are emphasised by other investors to such an extent that it has a significant impact on the pricing of these assets, this will, when taken in isolation, reduce our interest in such investments.

5.

CEM shows that the costs of managing unlisted assets are generally higher than those of managing listed assets, but also that there are large variations depending on the asset management model (internal/external, direct/fund constructions, etc.).

6.

The rate of home ownership by households in Norway is high compared to other countries. Moreover, households and small businesses account for a large portion of the supply side in the residential lettings market.

7.

If, for example, one were to invest 5 percent of the GPFN in real estate over ten years, it would involve investments in the region of about NOK 1 billion per year.

8.

Return on office properties in Oslo 1982–2013, Peter Hermanrud, Swedbank, 27 May 2014.

9.

This issue is analogous to monetary policy discussions concerning the importance of ensuring that the central bank keeps its promises, to create trust and reduce unwanted risk premiums.

10.

Available on www.ftf.no.

11.

See our strategic plan, available at www.ftf.no, for further discussion.

12.

An unlisted investment with a stable valuation, which can be said to reduce overall risk in the Fund, may for example result in a significant increase in expected tracking error due to changes in value captured by the benchmark index.

13.

By way of illustration: we envisage a “handful of employees” in addition to the current staff of about 50, who will be subject to the same values, administrative provisions and salary structure as the rest of the organisation.

14.

See the discussion in the Ministry’s letter of 27 June 2014 and Report No. 19 (2013–2014) to the Storting.

To front page