What Is a Good Investment Portfolio?

A board of trustees is responsible for developing an investment plan that is the best possible option for realizing the fund's goals, given constraints on the amount of risk, horizon, liquidity, and any other constraints. The previous discussion has focused on the steps needed for portfolio construction, which run the risk of overshadowing the real purpose of the process for a board. What exactly, then, is the best possible investment portfolio? We provide you with a list of criteria (also shown in Exhibit 8.3) for judging the quality of a portfolio and for comparing, scoring, or ranking the quality of different portfolios.

A good portfolio will help the board:

• Realize the primary investment objective over the relevant horizon with a high level of probability (fit for purpose). Is there a high level of probability that the portfolio will realize the primary investment objective over the relevant horizon? Is it fit for purpose? If this is not the case, it is time to rethink either the objective or the portfolio. This, for example, led the Norwegian Fund to increase its equity exposure in order to generate the long-term returns it targets.
• Realize potential secondary objectives such as environmental, societal, or governance goals (sustainable). There may or may not be other investment objectives. For example, the South African pension fund Government Employees Pension Fund (GEPF) has the dual objective of realizing return as well as contributing to the economic growth of the local economy.
• Do all of this at the lowest possible initial investment (effective). Is the return as high as possible within the set risk appetite; is the risk budget fully utilized? If the risk budget is not fully utilized, the fund might be setting an ineffective premium rate, i.e. charging a higher amount than necessary.
• Be robust in as many circumstances as possible (robust). Will the portfolio be robust in different economic and financial scenarios? Many risk premiums simply need time to be earned. It does not make sense to have an investment plan with a high allocation to equities, requiring a long period to earn the premiums, and then change the asset allocation in just a few years. So from the start, the number of assets and resulting asset allocation in the investment plan should be based not on one, but on a number of different scenarios.
• Have as much flexibility as possible for adapting when needed (flexible). Is the investment plan flexible over a relatively short term? If an investment plan advises changes in equities and bonds, these are usually easily implementable within a few months. But if a large part of the plan is based on shifts to illiquid assets over a longer period, then the board really needs to discuss how to commit itself to such future investment plans.
• Be based on a shared understanding of the drivers (acceptable). Is the investment plan clear about the main drivers and choices in the investment plan? What are the risk premiums, the amount of risk that is accepted to earn those premiums, the investment styles, and the horizon? If the board does not have a shared understanding of the main drivers, new board and committee members will challenge and rethink these drivers over and over again, especially in case of disappointing results. This reduces the horizon of the investment plan considerably.
• Involve the lowest possible cost (cost effective). Costs (meaning all costs, visible and less invisible) are always hotly debated. One school of thought argues that it is the net result that counts. Another school argues that costs are a factor that can be managed, and you should always work on cost efficiency because higher-than-necessary cost could be considered misspending the participants' premiums. This is the school we are in. There is quite a lot of evidence to substantiate the view that managing for lower cost is one of the most effective ways to increase the long-term return.²
• Have the continued support of the board and any other critical stakeholders to stay the course when things don't work out as planned in the short term (supported). Will the board uphold the investment plan even when things go sour? Suppose there is a period of financial turmoil: diversification between equities and bonds will probably work (equities go down, bonds go up) but all the other assets will move more or less in the same direction—downwards. In other words, every now and then, in practice, the quantitative assumptions about correlation, diversification and investment styles fail. Would a board still consider the investment plan to be robust, even if these assumptions fail?
• Be set out in the most transparent, most understandable, and most controlled way possible (controllable). Is the board able to monitor the main choices in the investment plan? The drivers of the investment plan should be attributable and visible in monitoring and reporting. If the potential success of the investment plan hinges on the credit spread and this is not reported, trustees cannot monitor or evaluate their own choices.
• Take the simplest, most elegant form in the most transparent, most understandable, and most controlled way possible (“parsimonious and elegant”). Is the portfolio as simple as possible, given the objective? Can it be made simpler? Often, complexity comes at a great cost, reducing transparency, oversight and understanding. Ceteris paribus, simple is better than complex.

When actually scoring the quality of a portfolio, it is a good idea to assign a weight to each of the criteria, thus helping the board to develop the portfolio that best meets its expectations, as well as providing a framework for a disciplined evaluation of it. For example, for some boards the ease of controlling a portfolio may have a high weighting, whereas other boards might give maximum weight to realizing the primary investment objective.

Among the criteria, there are definitely a lot of trade-offs that a board must consider. For example, trustees might expect to be able to generate another 0.5% annual expected returns at the “cost” of carrying out a number of complex and hard-to-control investments.

Diversification

Diversification is a much-used term within pension funds. Two questions are of importance for trustees. First, do the trustees share the same notion as to what diversification means for the fund? Secondly, do the assets chosen for diversification stand the test of time, are some of them perhaps only temporarily attractive, or are they truly new and do they diversify risk premiums?

Pension funds are clearly fond of diversification. Since 2002, they have ventured into credits, emerging markets equity, infrastructure, hedge funds, and private equity, while in recent years factor investing or impact investments have experienced an upswing in investment portfolios. Diversification, discussed in Chapter 4, can have different meanings to different people. Trustees should raise the question of which purpose or combination of purposes is relevant to their fund, as each approach leads to a different portfolio:

• Diversify to reduce overall portfolio risk. Whenever two imperfectly correlated assets are placed in a portfolio, there is an opportunity to earn a greater return at the same risk, or earn the same return at a lower risk. The correlation between the new asset and the assets in the portfolio should reflect a sufficiently different pattern of returns; and the allocation of the new asset should be sizeable enough to matter.
• Diversify to hedge against adverse asset pricing shocks. If there are extreme price movements in the financial markets, will the new asset's lack of correlation disappear and is diversification then be suspended? Or does the new asset provide a really new, independent economic source of return that is ex-ante unlikely to move in step with the other assets? This seems like a simple question, but it is actually hard to answer. For example, research shows that country equity markets offer less diversification in times of falling markets than in times of rising markets. The same is true for global industry returns, individual stocks, hedge fund returns, international bonds, etc. Similarly, international diversification seems to work during good times, when it is least needed, but to disappear during falling markets, when it is most needed. Maybe the only light beacon here is that asset correlation within countries decreases during periods of market turbulence.⁴
• Diversify to achieve high absolute and risk-adjusted returns. Is the new combination of assets able to increase outcome relative to the major asset classes in the portfolio? The appropriate measure to use is the Sharpe ratio, measuring the excess return above the risk-free rate per unit of risk.⁵
• Diversify to hedge against unexpected inflation or deflation. Conventional wisdom has held that assets such as real estate and long-term bonds hedge against inflation. However, with the exception of inflation-linked bonds, many assets have a complicated relationship with inflation. Inflation elicits differing responses; real estate contracts can have a standard clause in which rents are corrected for inflation, creating an adequate inflation hedge. Many assets by definition provide a partial inflation hedge; inflation may raise future cash flows, but this is partially offset, as inflation also raises the nominal interest rate to discount the cash flows, eroding the overall valuation. In addition, the fact that there is a historical correlation with inflation is not sufficient reason to extrapolate into the future. Unforeseen consequences can always affect the future. For example, thanks to the enormous success of Wal-Mart, few retail chains were able to pass on increased inflation to customers; Wal-Mart alone is credited with reducing inflation in the United States by one percentage point per year in the 1990s.⁶
• Reflect the overall investment universe. A portfolio with 50% equities and 50% bonds might be called a balanced portfolio, but it is definitely not a reflection of the investment universe. This implies—in theory—that any portfolio that does not include private equity for example is making an implicit bet that a portfolio without private equity yields better returns.
• Diversify to deliver a stable stream of cash flows to the investor. Pension funds are greatly concerned with liquidity, a major concern being the creation of a stream of cash flows that to a large degree offsets the pension payments to be made. Investments with strong cash flows provide a natural hedge against pension liabilities, which are basically a set of guaranteed future cash payouts.
• Diversify to capture the underlying economic and financial factors that drive asset returns. Taking a quantitative approach, equities or bonds are not viewed as issued by companies, but rather as representing a bundle of risk factors. According to this view, “decomposing” these assets into risk factors allows the investor to diversify far more effectively.

How many assets are needed for diversification? Having come to a common understanding of diversification, do the assets chosen for diversification stand the test of time, should the set be expanded, or are just a few assets sufficient for effective diversification? In allocating the assets, regulators expect the trustee to diversify the holdings and to exercise prudence. In other words, trustees should not bet the house on anything too risky in the portfolio.⁷ However, views on diversification have changed dramatically in the past decades. Pension funds in the 1970s were content with bonds and real estate, assets that were considered to be “safe.” But returns were low, so funds progressively moved into equities, commercial property, foreign securities, and derivatives. New technology made this possible. In the 1990s, the pension fund industry experienced a makeover when the combination of Asset Liability Management (ALM) techniques, deregulation, and portfolio optimization approaches gained ground, widening the scope of assets to invest in. The risks were higher, but so were the rewards. Additionally, higher returns brought down the relative costs of running the fund.

In hindsight, it appears that only a limited number of strategies break through and help to diversify effectively. In other words, trustees should quiz their investment managers and advisors critically, because introducing new strategies does not necessarily achieve diversification. Alternative investment strategies have suffered from two problems. First, the success of many new strategies was boosted by the same factors: low interest rates and robust economic growth. This, in turn, encouraged investors to use leverage to enhance returns. Some alternative investments share more factors; private equity gives investors exposure to the same kind of risk as publicly quoted equity, albeit with added leverage. When prospects deteriorated, investors were forced to sell all those asset classes simultaneously. Secondly, some of the asset classes are rather small. This illiquidity is attractive since it offers higher returns. As more investors get involved, the market becomes more liquid, and the higher return is eroded. However, when everyone tries to sell, illiquidity rears its ugly head again—there are no buyers to be found and prices tumble. This is very unpleasant when pension funds have to value their investment at market value. Still, investment managers increasingly look to add new—alternative—investments to bolster diversification advantages,⁸ which can mitigate these risks to a large extent. However, there is no point in adding new strategies if the investment does not offer a genuinely different source of return, or if the asset is already overvalued. In other words, the bar for new investments to count as a true “diversifier” has been raised substantially, and if these investments are true diversifiers, their relative size in the investment portfolio will be limited.

Along with the search for alternative investments as a source of diversification, factor investing has gained traction, viewed by some as a replacement for the current approach to diversification. Interest in factor investing started with the Ang, Goetzmann and Schaeffer study,⁹ commissioned to investigate why active returns of the Norwegian Fund were disappointing. A major finding was that a large share of active returns could be attributed to systematic factors; five years later, major pension funds such as the Alaska Endowment Fund, the Dutch PFZW or the Danish Arbejdsmarkedets TillægsPension (ATP) all implemented a form of factor investing. From the outset, it seemed like a sensible development. Asset returns can be attributed to a few distinct underlying factors such as macroeconomic factors, firm attributes or style factors, with these factors representing sources of systematic risk and return. This approach aims at providing a relatively precisely defined exposure to factors that may explain differences in asset returns and, to a large degree, the excess returns of actively managed portfolios. The factor approach creates the opportunity to construct portfolios with asset selection based on the knowledge of underlying factors, or even design portfolios using factors rather than assets.¹⁰ Within this factor framework, which is rooted in the Asset Pricing Theory (APT) developed in the late 1970s, any asset can be viewed as a bundle of factors that reflect different risks and rewards. Therefore, it makes perfect sense to focus on these factors, allowing investors to identify them and understand what really drives asset returns. These insights help to develop portfolios that realize the required risk profile not only before, but also during, volatile periods.

Rebalancing

Rebalancing is regarded as a technical component of implementing an investment portfolio plan, where portfolio weights are adjusted in line with the risk appetite of the fund. This policy details when and how the policy portfolio is adjusted in line with changes in the cover ratio or financial markets, how to adjust the assets to the weights of the strategic benchmark portfolio, and whether the rebalancing is done on a frequent regular basis or based on specific triggers.

Rebalancing on the basis of constant policy weights represents one possibility. At the other extreme, the fund can have a solvency program in place, which adjusts the level of policy weights depending on the amount of downside risk the fund is willing to accept given the funding ratio and financial market volatility. Similarly, the solvency program adjusts the level of policy weights depending on how close the fund is to achieving its long-term goals. Sometimes, funds allocate part of the bandwidth around policy weights to TAA (tactical asset allocation), which involves making short-term adjustments to asset-class weights based on short-term expected relative performance among asset classes. The assumption is that investors can take into account short-term opportunities with the aim of achieving the best performance for the portfolio, bearing in mind the level of risk defined.

At times, a portfolio's actual asset allocation may purposefully and temporarily differ from the strategic asset allocation. For example, the asset allocation might change to reflect a pension fund's current circumstances when these are different from normal. The temporary allocation may remain in place until circumstances return to those described in the investment policy and reflected in the strategic asset allocation. If the changed circumstances become permanent, the manager must update the investor's IPS, and the temporary asset allocation plan will effectively become the new strategic asset allocation. TAA also results in divergence from the strategic asset allocation, and responds to changes in short-term capital market expectations rather than to investor circumstances.