Using as many alpha sources as you can
The Norwegian Ministry of Finance has stated that it wants the Government Pension Fund, the umbrella scheme that includes the former Petroleum Fund, to become the best managed pension fund in the world.
The problem with this kind of aim, as ministry officials concede, is that there is no real peer group against which the performance of the fund can be judged. In terms of size, the fund ranks among the largest European pension funds with assets of €200bn . Yet it operates more like an endowment than a pension fund, since its objective is to pay out the return on its capital rather than the capital itself.
The name of pension fund, which it was given on 1 January 2006, is also slightly misleading. It was established in 2005 by an act of parliament as a superstructure spanning what was previously the Government Petroleum Fund and the National Insurance Scheme Fund. These two entities have been renamed, respectively, the Government Pension Fund - Global and the Government Pension Fund - Norway.
The renaming of the Petroleum Fund followed the report of the Pensions Commission, which felt that a specific reference to pensions would underline the fact that the fund would be an important element in Norway's strategy to meet future pension expenditures. Yet currently the real return of 4%, which it is expected to generate annually, is intended to match the budgetary deficit in the non-oil economy.
eanwhile the two broad objectives of the Petroleum Fund remain - to avoid extreme cyclical fluctuations in the Norwegian economy due to variations in oil revenues, and to ensure a fair distribution of oil wealth among generations.
The Norwegian Government Pension Fund has become a model for other resource-rich countries, who send delegations to Oslo to find out how the Norwegians have managed transmute their oil and gas revenues into foreign financial securities.
The model has succeeded principally because there is a clear division of responsibility between the Ministry of Finance, which is the owner of the fund and Norges Bank, which is the operational manager.
The return of the fund is determined by a strategy decided by the Ministry of Finance. The strategy is embodied in a benchmark portfolio of equity and bond indices. FTSE indices are used for equities and Lehman indices for bonds. The benchmark portfolio is made up of 40% equities and 60% fixed income. The allocation, to equities has remained unchanged since it was introduced into the portfolio in 1998.
Norges Bank is responsible for implementing this strategy and producing the ‘value added' against the benchmark. The job of adding value to the benchmark has been given to Norges Bank Investment Management (NBIM) the asset management wing of Norges Bank created in 1998.
NBIM's asset management activities can be divided into two. The first is the beta function - ensuring that the fund has full exposure to the market at the lowest cost. The second is the alpha function - looking for excess returns. It is this function that represents the greatest challenge, says Knut Kjaer, executive director of NBIM.
The active management of the fund has evolved over a number of years, says Kjaer. "We started very carefully, with the belief that the markets are nearly efficient and that to beat them you need special skills. We did not try to take big positions, or to be everywhere at once. Instead we used the early days to see where we could realistically beat the market."
Kjaer and his team eventually found their approach to active management in a classic textbook - Richard Grinold and Ronald Kahn's ‘Fundamental Law of Active Management.' Grinold and Kahn's law states that the best way to add value in an active portfolio is to make as many independent bets as possible.
Since the fund began to invest in equities in 1998, active management has produced an average annual excess return of 0.5 percentage points, with no increase in the risk of the portfolio. It has achieved this with an information ratio of 1.3. The ratio, which shows how much return is achieved for each unit of risk, is the key measure of the skill of active managers.
Kjaer has distilled the message of Grinold and Kahn into an elegant equation. "The chances of getting a high information ratio is a function of your information coefficient - your ‘hit ratio' or how well-informed you are in the positions you take - and the square root of the number of your positions."
Having a multiplicity of well-diversified sources of alpha is crucial, says Kjaer. "We see the probability of having successful active management is a function of the ability to find as many independent alpha sources, and the specialists in the skills at all those alpha sources."
Kjaer does not mind where he finds these, internally or externally. Although the proportion of the portfolio managed by external managers has decreased over the years, this does not reflect a move to take all management of the portfolio in-house, Kjaer says.
"I don't have a master plan of building a big internal asset management organisation. Our challenge is to find more alpha talents, for both internal and external management. We have so far very positive experience from building up the internal alpha capabilities and benefit here from being located also in London and New York."
he cost of using external managers is considerably higher than the cost of using internal managers - 31 basis points compared with 4 basis points. Yet the excess returns external managers can produce more than cancel out their costs, says Kjaer. Although external managers represent 62% of the fund's management costs they take the same percentage of the fund's active risk.
"It's not about getting the cost down. It's about finding the net-of-fee, good alpha sources. So all the time we can see improvement over expected excess return we will use the opportunity," he says.
The performance of managers is monitored closely and regularly, Kjaer says. "We see everything they do to the portfolio, their transactions and positions, on a day to day basis. We have systems to track both the portfolios and the combination of portfolios and their impact on the total portfolio.
"What we use this data for, and what we look at when we consider new managers, is every alpha component. Is there a consistency between the investment philosophy, what is written in the portfolio, and what is actually being done in the portfolio? Then there is the hits ratio. Is last year's excess return the result of one big decision or of 100 big decisions?"
Turnover of external manages averages 10% a year. The main reason that mandates are terminated, Kjaer says, is the movement of managers.
"Turnover is too often because of key people in the investment management companies leaving.," he says. "In too many cases we had to terminate a contract was because of a company losing a key person running the portfolio." The last termination was costly but justified, he says. "We took on costs that could be measured in hundreds of millions of kroner, but it's better to take that early on rather than later, when all the other clients are pulling out." Kjaer puts a high value on talent. "Since we started with external management in 1998, our experience has been that talented people are much more important than we believed in the beginning.
"We regard successful alpha products out there as very much linked to talented people, not linked to a brand name, a big organisation, good systems, or a good track record. What we buy is future excess return. We don't buy history. The future is very much linked to talented people staying in organisations."
The search for alpha is only one, albeit the most important tasks of NBIM. Keeping transaction costs as low as possible is another. The Norwegian government transfers between $20bn and $40bn in oil and gas revenues into the fund each year, and the job of moving cash into equities and fixed income seamlessly into the fund is a major challenge, "Transitioning in itself is a really important business for us and we put a huge emphasis on keeping the transaction costs down," says Kjaer.
Norges Bank uses all the tools at its disposal, such as electronic crossing networks, to enable it gain as efficient access to the markets as possible, he says. Last year it established a direct link with world stock exchanges. One area where NBIM is determined to improve cost transparency is broker research. Next year, NBIM will start to buy external research and add the cost to its budget.
"The issue of bundling services has been our agenda for some time. We as a big client of the investment banks of course recognise that we are subsiding their research to smaller clients," says Kjaer. "So we are working on getting good efficient systems for compensating the investment banks, putting the execution cost where it should be and paying the right price for research." Another ongoing task of NBIM is ensuring that the organisation it has built from scratch remains fit for purpose. The development of active management skills in-house has much to do with the way the investment manangment operation has been structured. Kjaer says he was anxious that it should develop as a business unit rather than a public
sector bureaucracy. He was also determined that individuals, not committees, should take decisions.
As a result NBIM does not have investment committees and ‘house views' of market behaviour. "We don't have a top department or a boss like me saying the market is going in a certain direction and instructing managers to follow," Kjaer says. "We don't even dictate the investment process of the internal managers. We provide resources and systems, access to company information and research, and we define clear long-short mandates and monitor the performance. But our people can decide themselves what investment process they want to apply. We never second-guess them, and the reward structure is clear cut and mostly based on individual performance."
Portfolio trading decisions are delegated to more than 70 external portfolio mandates and to 15 internal teams, and within these teams to individuals. The aim is to avoid collective decision-making, or ‘group think'. "We try to avoid committee decisions and we try to empower people to take investment decisions." he says.
The flat organisational structure produces a ‘lean and mean' operation that enables managers to move quickly in and out of investment opportunities whenever they see them, he says.
Kjaer and his chief investment officers can still take the tiller if they need to. "If we see there's a real investment opportunity, we have systems for just going and taking the benefits of that opportunity. But these are likely to be decisions of the chief investment officers. It could be a decision of mine, but my threshold for doing that is very high."
NBIM has a reward structure that reflects this empowerment. This is also reflected in the external management. By delegating responsibility for investment decisions to in-house investment officers, Kjaer makes them responsible for the success - or failure - of the external managers they have chosen.
"Everyone is made responsible for their return and their salary is related to that return," he says. "I compensate my people also on the external management part of this business based on the performance of external managers."
Keeping all these initiatives going is taxing work, says Kjaer. "The most challenging part is combining transitioning, beta management, getting excess return and building our organisation from scratch. Conditions for creating excess return have been different from other funds because we have had to build the portfolio and the organisation at the same time."
All this takes place within a framework administered by the Ministry of Finance. A tracking error of 1.5% ensures that excess returns are earned at a reasonable level of risk, and a re-balancing system ensures that the fund's portfolio stays close to the strategic asset allocation.
This ‘smart" rebalancing channels monthly inflows of new capital into those asset classes or regions with the largest negative deviation from the benchmark. This mechanism played a significant part in enabling the fund to recover from the fall in equity prices in 2001 and 2002. It ensured that while others sold, Norges Bank bought.
"Having the rebalancing system meant that we bought heavily into the equity market in 2001 and 2002 and avoided the catastrophe of selling equities at the bottom of the market," says Kjaer.
This episode may have reinforced the case for increasing the fund's exposure to equities. Initial funding, which began in 1996, was invested in government bonds. New regulations in 1998 allowed investment in equities within a limit of 30% and 50% of the fund. The benchmark portfolio's equity portion was set mid-way at 40%.
The 40% allocation to equities survived the collapse of the equity markets in 2001-2002, and there is now a move to raise the 40% limit. Norges Bank has recommended raising the portion to between 50% and 60% while the Investment Advisory Council, a body of experts that advises the Ministry of Finance, has recommended 60%. The Ministry of Finance has yet to decide on the issue.
The clear division of responsibilities between the Ministry of Finance and Norges Bank in the running of the Government Pension fund - Global is also apparent in the areas of SRI and corporate governance. The Ministry of Finance is responsible for the SRI function - specifically the filtering and exclusion of companies - while Norges Bank, as the primary owner of the assets, is responsible for corporate governance. There is now a move to increase the bank's activity in the area of governance. Last September, NBIM set up a corporate governance operation headed by Henrik Syse. "According to our mandate and guidelines we are to be active owners in the portfolio," he says. "We have been active in proxy voting for some years but we are relatively new in raising corporate governance to a high level in the bank.
"It is long term work and its something where you need an ownership horizon to do it. That creates a seriousness and thoroughness for our corporate governance work. Because when companies see us raising issues they know this is not some flavour of the month but a serious concern."
This is - and is meant to be - somewhat different from the intention of Ministry of Finance's screening and exclusion of certain companies, measures that are not intended to be ‘punitive' but to remove the fund from the risk of complicity. "The Minister of Finance, as our principal, describes the scope of our universe by deciding the benchmark and also by deciding those companies that we can not invest in for ethical reasons," Syse explains.
"Part of the reasoning here is that those decisions are made not primarily on a financial basis but because we should not become complicit in gross violations of basic ethical standards. That is closer to a political decision than an investment decision."
Exclusions, which have totalled only 19 to date, have had a minimal impact on the diversification of the portfolio, says Syse. "Quite consciously the threshold had been set high in order not to destroy the general profile of the fund, which is that it is a broadly invested fund managed according to certain risk parameters. But exactly because of that, we need to be active owners and voice our concerns through engagement with companies and standard setters. Those concerns include social, environmental and governance issues." The growing importance of the Government Pension Fund - Global was also flagged up last September when the Ministry of Finance created a dedicated asset management division. Strategic asset management of the fund was formerly the responsibility of a section within the ministry's economic policy department. The director-general of the new division is Martin Skancke, a former director general of the office of the Norwegian Prime Minister and someone who was closely involved in the setting up of the Petroleum Fund in 1996.
"The creation of the department reflects the fact that fund has become very large and the operation very complex.," he says. " We are trying to build an environment where there is sufficient leadership attention to the fund, where it's not just one of many issues in one department and where it is actually singled out as an important area in its own right."
Skancke says the Government Pension Fund - Global is now seen as a possible model by other countries. "There is a clearly high demand internationally for a technical support and transfer of knowledge to other countries. We work extensively with the IMF and the World Bank to disseminate knowledge on how to set up petroleum funds and manage petroleum revenues. "We don't pretend that it's easy to transfer or apply our experiences to other countries directly, because we were fortunate enough to have at the outset well-functioning institutions, transparent government, and we had a well-functioning central bank which was well equipped to set up a fund management operation. One feature of the Government Pension Fund which is transferable is its transparency, he suggests. "It is vital for the success of the fund that we have this high degree of transparency. If you ask people to trust you to invest in what is essentially their money you had better be prepared to tell them exactly how you are going to do it."