5 Review of Folketrygdfondet’s management of the Government Pension Fund Norway

Letter of 16 December 2014 from Folketrygdfondet to the Ministry of Finance

We refer to the letter of 27 June 2014 from the Ministry of Finance, in which Folketrygdfondet is requested to submit analyses and assessments concerning the implementation of the management of the Government Pension Fund Norway (GPFN). As background materials supplementing this letter, we will refer to the following two enclosures: (a) Strategic plan for the management of the GPFN and (b) Analyses and assessments of performance in the management of the GPFN. We find that the current framework for the management of the GPFN is appropriate, and will not be proposing any changes in this letter, although we will refer to a separate letter of 13 November 2014 concerning unlisted real estate and infrastructure investments.

Strategic plan

The strategic plan for the management of the GPFN has recently been adopted by the Board of Directors, and will be published on our website before the end of the year. The plan describes the strategy of Folketrygdfondet for the management of the GPFN, and is premised on the mandate laid down by the Ministry of Finance as well as on the distinctive features characterising the GPFN and Folketrygdfondet. These distinctive features include our governance structure, long time horizon, capacity to absorb risk, limited liquidity needs, size and experience.

Folketrygdfondet seeks to be a responsible owner and investor, adding value to companies and promoting well-functioning markets. Folketrygdfondet aims to achieve high returns over time for the GPFN by increasing overall returns in the market and by generating excess return relative to the general market.

The excess return is to be generated along several dimensions in line with our distinctive characteristics and advantages as a manager, and risk taking will vary over time depending on investment opportunities. We believe that said characteristics and advantages, our values and investment philosophy, as well as our extensive experience provide a solid basis for continuing our active management of the GPFN and for generating excess return at low operating costs. Active management is more suitable for the GPFN than passive management because:

• Liquidity in the Norwegian financial market is variable and at times poor.

• The risk exposure implied by the benchmark index is not always suitable for a long-term investor.

• There is mutual interdependence between responsible investment and active management.

In addition to exploiting specific low-risk opportunities for creating positive returns, which arise from the special characteristics of the portfolio, we would like to highlight the following sources of excess return as being of importance in both fixed-income and equity management:

• We select quality companies in which we are willing to invest with a long-term perspective.

• We reap liquidity premiums.

• We utilise the fact that risk premiums, and hence expected returns, vary over time.

Historically, we have emphasised that risk taking shall be dominated by company-specific matters in the equity portfolio, and by general credit and liquidity risk in the fixed-income portfolio. We will maintain this perspective, whilst at the same time emphasising the need for the strategy to evolve in coming years, in order to ensure that our investment activities are tailored to new challenges and opportunities. New activities shall, as a general rule, be derived from our distinctive characteristics and advantages, and it is our position that Folketrygdfondet shall remain a cost-effective asset manager.

Folketrygdfondet has, over time, generated considerable added value in our asset management relative to the benchmark index. This gives us confidence in the strategy for the management of the GPFN which has been pursued over many years. We also find support for key elements of the strategy in academic studies. In our strategic plan, we seek to describe how the strategy can be linked to so-called systematic risk factors, as identified in empirical studies.1

As regards our estimates of expected average excess return and risk taking (as measured by tracking error), we would like to highlight the following:

• We have achieved good active management performance over time. The historical excess return is slightly higher than, but consistent with, the established objective of an annual excess return relative to the benchmark index of 0.4 percentage points (before asset management costs).

• Since investment opportunities and expected returns vary over time, our risk taking must also vary over time. Our active risk taking has declined in the wake of the financial crisis, and is currently at a level that is indicative, when taken in isolation, of an excess return that is significantly lower than the objective. We continue to assume that new investment opportunities will materialise in the longer run that will provide a basis for risk taking and excess return in line with the established objective.

Return and risk

The enclosed report “Analyses and assessments of performance in the management of the Government Pension Fund Norway” analyses return and risk in the GPFN, as well as for the equity and fixed-income portfolios individually. The analysis examines the years 1998 to 2014.2 The main emphasis in the analysis is on the period after the mandate was modified, i.e. the period 2007–2014, and on the last four years, i.e. the period 2011–2014. All figures we quote here are calculated as time-weighted returns, and specified as annual geometric averages. Excess return is specified as the difference between the annual return on the portfolio and the annual return on the benchmark index. We would like to highlight the following:

• Annual return was 7.4 percent for the period 1998–2014 as a whole.

• Annual excess return was 1.1 percentage points for the period 2007–2014, and 0.5 percentage points for the period 2011–2014. This is better than the established objective of an annual excess return equivalent to 0.4 percentage points.

• The data on active management show that excess return has been positive for all three time periods, both for the GPFN as a whole and for the two sub-portfolios.

• The risk-adjusted excess return shows that Folketrygdfondet’s active management of the GPFN has served to improve the ratio between return and risk, as compared to the benchmark index.

• The analysis shows that the excess return can largely be explained by elements we have identified as being key features of our strategic plan for the management of the GPFN.

• Analysis of the exposure to systematic risk factors conducted by way of regression analysis indicates that the excess return is largely generated through the selection of individual securities, and only to a lesser extent through exposure to systematic factors. The chosen method estimates the effect of a constant exposure, whilst exposure to systematic factors in the portfolio has varied over time. We have identified time-variable exposure as an active management tool in the strategic plan.

Performance in the active management of the GPFN has been favourable, both for the GPFN and for the equity and fixed-income portfolios.

Costs and value added

The costs incurred by Folketrygdfondet in the management of the GPFN increased relative to assets under management over the period from the modification of the mandate in 2007 until 2009. This is primarily the result of higher system costs relating to portfolio and risk systems. Over the period from 2009 to 2014, costs have varied between 8 basis points and 10 basis points of the GPFN capital. This is a very low cost level compared to other asset managers; see the calculations prepared by the Canadian firm CEM Benchmarking at the request of the Ministry of Finance.

Value added in active management and gross differential returns cannot be compared directly. However, our analyses of relevant cost and income components for passive management indicate that gross excess return is also a good indicator of net value added from active management. These analyses show that the generally low costs associated with management of the GPFN would not have been very much lower in the case of purely passive management.

Risk management

The enclosure “Analyses and assessments of performance in the management of the Government Pension Fund Norway” provides a description of Folketrygdfondet’s risk management. The description includes the following observations:

• Folketrygdfondet adheres to the Regulations relating to Risk Management and Internal Controls (“Risk Management Regulations”) laid down by the Financial Supervisory Authority of Norway, to the extent applicable.

• The Board of Directors of Folketrygdfondet adopts risk management principles, the investment mandate for the Chief Executive Officer, a strategic plan for asset management, responsible investment principles, a job description for the Chief Executive Officer and employee remuneration principles.

• The Chief Executive Officer adopts guidelines setting out more detailed risk management and control requirements, and issues written authorisations in relation to the investment activities.

• All managers have risk management responsibility within their area of responsibility and authority. Risk assessments are an integral aspect of the business processes, and also encompass outsourced services.

• Financial risk limits are monitored continually through the monitoring systems of Folketrygdfondet, and any incidents are registered in a register of incidents. Any breach of limits stipulated by the Ministry of Finance or the Board of Directors is published in the annual report of Folketrygdfondet and the quarterly reports on the Government Pension Fund Norway.

• Annual discussions are held with the Board of Directors to evaluate the risk management, return measurement and internal control systems of Folketrygdfondet.

Passive management

The Ministry has requested our assessment of potential consequences of more passive management of the GPFN. Public debate often tends to leave the impression that passive management involves less investment risk and lower asset management costs than active management. This perspective needs to be nuanced.

A model in which the capital owner defines a benchmark index clarifies the willingness of the owner to take risk and makes it clear what responsibility lies with the capital owner and the asset manager, respectively. However, as previously argued in this letter and outlined in more detail in the strategic plan, we are of the view that there are several reasons why it would not be prudent to invest the GPFN rigidly in conformity with a benchmark index that is not necessarily tailored to the distinctive characteristics of the GPFN. Reducing the responsibilities of Folketrygdfondet and opting for more passive management of the GPFN will, when taken in isolation, increase risk in the portfolio (less diversification of risk), reduce the return net of costs, contribute to a less well-functioning financial market (large market effect when index changes take place in a capital market that is, at times, illiquid) and reduce the impact of responsible investment activities. Consideration for the reputation of both the capital owner and the asset manager suggests that any transition to more passive management should be thoroughly analysed and clearly justified in accordance with the internationally recognised standards on which asset management is otherwise premised. Against this background, we would advise against passive management of the GPFN.

Yours faithfully

Enclosures (available on the Ministry website in Norwegian only).

• Strategic plan for management of the GPFN

• Analyses and assessments of performance in the management of the Government Pension Fund Norway