Irrational despondency follows exuberance
Pension managers are yearning for the old days, when they only had to worry about benchmarking and tracking errors. Today there are a variety of other issues to deal with, ranging from new regulations to yawning funding deficits. Such lean times, especially when they follow the fat years that pension managers enjoyed not so long ago, have many looking for a scapegoat. Indeed, whole classes of investments, particularly equities, have come under fire for yielding nothing but losses and acid indigestion for pension managers.
The current rejection by some of all things equities is no more rational than the exuberance of the bubble at the end of the last decade. No one asset class is inherently good or bad. Equities, fixed income, real estate, hedge funds – they simply are what they are, all with their own strengths and weaknesses. It is how and in what proportion these asset classes are combined that makes all the difference.
Now, with the painful clarity of hindsight, managers can see that, during the halcyon late 1990s, they were far too exposed to certain classes of equities – particularly US large-cap growth funds. Our own experience as asset managers supports this claim. During the bubble, we were approached by several potential clients seeking an unbalanced investment structure. For some of them, their only exposure to US equities was in discrete segments of the US market. A number of institutions that approached us owned only S&P 500 index funds, yet believed they owned the US market.
The responsibility for imbalanced portfolios does not just lie with pension funds. Asset managers bear their measure of responsibility as well for selling one type of strategy and through chasing performance, delivering something entirely different. During the bubble, balanced strategies drifted towards growth and value strategies drifted towards core. The growth strategies just got growthier. There was also drift across size, all converging on the class of equities that was booming. As a result, even if pension funds knew better than to lump their holdings into one class of fund, they may have ended up more concentrated than they realised, until the bear market made it all too clear. Imbalances in equity holdings left pensions more exposed than necessary to the huge decline in equity prices over 2001 and 2002. From 2000–02, pensions lost over $500bn (e450bn) in the US alone. British pensions lost nearly £1trn (e1.4trn).1
With these dramatic losses, it is unsurprising that the popularity of equities has waned. Initially, the asset classes that benefited from the allocation shifts were safe havens, such as bonds and money market instruments. Famously, the Boots Pension Fund in the UK adopted a 100% allocation to bonds. More recently, alternative investment strategies, such as hedge funds, have also capitalised on the dramatic policy changes, positioning themselves to the disillusioned investment community as a possible ‘third way’, with the promise of high returns and low correlation to the public markets – an attractive proposition for nervous managers facing growing liabilities, expanding fund deficits and a troubled stock market.
After the brutal beating of the past few years, it is no wonder that some managers are wary of any new investment in equities. However, simply writing off equity investments is counterproductive. The great irony for pension managers is that the best way out of the funding deficits is to embrace the asset class that produced the hole in their pension funds in the first place. Equities are the only class of investment with the long-term track records for returns that can erase the deficits. However, as recent history warns, equity investments must be diversified and balanced.
In the beginning…
The appeal of equities is of course well known, but it is worth reiterating. On the most basic level, investment in stocks is an investment in the future earnings potential of businesses and, more broadly, the economy as whole. Equities are therefore a useful hedge against inflation. From a pension fund perspective, this can help deal with issues such as wage drift.
By contrast, nominal bonds (historically, the other major component of diversified portfolios) provide a hedge against deflation – offering a fixed return over a fixed term. A balanced portfolio with both types of assets can, therefore, normalise the returns of a portfolio whatever the inflationary environment. Regrettably, with emotions running high over the past few years, this simple logic seems to have eluded a large number of market participants.
Additionally, stocks offer investors the possibility of high returns capable of making up deficits and creating surpluses. This potential for uncapped upside is one of the fundamental differences between nominal bonds and equities. Needless to say, this potential for upside is matched by a potential for downside, but in the current environment of growing liabilities and large deficits, the need for returns has left even the most risk-averse pension fund managers with little option other than to tilt their portfolios towards a riskier bias.
Equities historically outperform bonds, and for those companies with large funding requirements or deficits, equities may be the surest way to make it up. According to Figure 2, the mean return for equities during the last century was in excess of 9% – a significant premium over bond returns that seems to justify the extra level of risk inherent in stock investing.
However, critics of equities believe that using this long-term track record to project future mean performance for stocks is misguided. They argue that the dramatic events of the past hundred years – including two world wars, high inflation and globalisation of the business landscape – created an environment uniquely suited to equities. They also maintain that visibility for similar inflexion points in the current century is low and, with e-commerce gradually eroding business margins in many market segments, we should expect more modest returns from stock investing than what has historically been achieved.
If the 100-year data is somehow biased, perhaps shorter time periods would be more informative? A study of the UK equity market by ABN Amro and the London Business School, A Century of Investment Returns, showed that double-digit returns from equities are a relatively recent phenomenon. From 1980– 2000, the annual real rate of return was 13.3%. However, the two decades prior to this (1960–80) generated an annual real rate of return of only 2.5%.
The start of the century saw returns of 0.2% for 1900–20 and only 5.9% for the period 1920–40. The post-war boom gave higher returns of 8.3% from 1940–60, but still below the levels seen at the end of the century. For the 50-year time period, meanwhile, the two halves of last century told two very different stories: a 3% return for 1900–50; and 8.9% for 1950–2000. In the first 50 years, returns clearly did not provide investors with adequate rewards for the risk of losing their principal. However, the second half of the century generated attractive levels of return.
Unfortunately, this closer look at the numbers arguably raises more questions than it answers. Elements of the current environment are common to a number of the periods listed, which makes choosing between them a fruitless and subjective exercise. If mean regression is inevitable over the longer term, then surely the 100-year numbers are still the best proxy available? Without the aid of a crystal ball, however, we are left with little option other than to use past performance as a proxy for future returns.
Inflation: which way?
Over the past 20 years, fund managers enjoyed the benefits of disinflation, which has been good for both bonds (which benefited from steadily falling yields) and stocks (which also benefited from lower interest rates). Now, inflation rates in most developed economies have nearly hit zero and further disinflation without deflation is almost impossible. In the case of deflation, demand for bonds with any sort of coupon should intensify while demand for stocks will wilt (see Japan for an example of this behaviour). Clearly, the deflation scenario argues strongly against future investment in equities.
If, on the other hand, the global economy is on the verge of reflation, the massive run that bonds have seen has come to an end and stocks look favourable again. If you buy the recovery story (and at some point it will almost certainly have to be right), then interest rates will tick up and the net present value of pension liabilities will go down. However, if a portfolio consists entirely of fixed income, the managers will find their ability to meet their obligations are reduced as well. Equities would be the more appropriate and responsible choice in a reflationary environment.
The point is, we cannot know with absolute certainty whether the future will bring deflation or re-inflation. But what we can say with reasonable assurance is that either outright deflation (as opposed to the disinflation of the past 20 years) or reflation will bring the correlation of returns between equities and bonds closer to zero (Figure 4). Therefore, it makes sense to diversify into both stocks and bonds to hedge against loss.
Even if you are not completely convinced by the recovery story (and there are plenty of legitimate reason not to be so sure) or even if you believe that interest rates will remain relatively steady at the current low levels, it must surely give investors pause that interest rates are at near 50-year lows.
When assessing the returns offered by equities, no discussion would be complete without a look at dividend yields. The aggressive gains in share prices during the strong growth period of 1980–2000 meant that income derived from dividends became relatively insignificant within the overall picture of returns realised from equity-investing. In an environment of leaner returns, however, dividends cannot be dismissed, and equity investors have been encouraged by the willingness of some companies (especially in growth sectors) to compensate investors for lower rates of growth in share prices by awarding dividends. At the end of the first quarter of 2003, approximately 50% of the S&P 500 had a higher dividend yield than the three-month US treasury bill – and that’s just dividend yield, not counting capital appreciation. The dividend yield on the S&P is not that far below the yield on the 10-year.
Critics of equities
Not surprisingly after three years of often crushing losses in the equity markets, critics have emerged and are questioning the prominent role that equities have traditionally played in pension fund allocations. There are a variety of arguments, ranging from cranky complaints about recent losses to valid challenges to traditional assumptions.
It has been well documented in these pages as well in other publications that companies have been gradually shifting their pension plans from defined benefit to defined contribution. But the impact of the decision may have had unforeseen consequences. By closing defined benefit plans to new employees, these plans face growing current liabilities, an ever-ageing workforce approaching retirement age and no new workers to make contributions. As a result, the time horizon for investments to meet obligations has shortened considerably. While equities may bring greater returns over the long run, critics claim that over the short term, when most of the obligations come due, fixed income is a more appropriate and less volatile investment.
As investment advisers, we have always steered such companies into more appropriate investment vehicle that better matched their liabilities. For companies with extremely short time horizons, equities may not be an appropriate investment. However, it has been our experience that such companies are a distinct minority in the market; especially in Europe where the commitment to defined benefit plans remains strong. If one intends to continue a defined benefits plan, virtually the only responsible way to yield the necessary level of returns to match future obligations (not to mention cost of living adjustments) is through equity investment.
Employees, the ultimate beneficiaries of pensions, may also prefer an equity allocation due to ‘benefit leakage’. If equities do well, then extra benefits may be passed along to the employees. However, if equities do poorly, then employees are still guaranteed their benefits and it is the responsibility of the company to meet these obligations by drawing on other resources. Therefore, employees share more on the upside and less on the downside. In countries or companies where labour has a strong voice in pension allocations, equities certainly have an advantage. The exception is of course if equities perform horrendously and the fund is pension is essentially bankrupted. Fortunately, such occurrences are rare, even over the past three years.
A second frequently heard criticism is that as the size of the assets under management by pension funds grows relative to their parent companies, the greater the impact on the company’s income. Since pension funds are becoming so large, returns just a fraction of a percentage point above or below the estimated return on assets equate to a sizeable impact on earnings in currency units. As return on equities is more volatile than fixed income, critics argue for reduced allocations to equities.
There is some validity to this charge. Furthermore, the movement towards mark-to-market accounting detracts from the ability of pension funds to smooth out the normal ups and downs of equity returns. However, there remain a variety of ways to hedge against the volatility of equity returns, such as derivatives, while preserving the long-term high yields that equities provide. A qualified asset manager can construct customised portfolios that can minimise risk while preserving the higher returns that pension funds demand to meet their obligations. Meanwhile, in many European countries where industry-wide pensions are the norm instead of company-specific funds, this simply is not a relevant consideration.
The problem of volatility and matching liabilities leads some managers to place greater emphasis on absolute returns and escape the tyranny of benchmarking. Such an approach starts from the perspective that a pension fund should first and foremost meet its liabilities and that the quality of asset allocation should be judged by how much it beats its liabilities and not some ‘arbitrary’ benchmark. Among the advantages to such an approach are that it reduces the temptation of becoming too concentrated in any one class of assets simply because there are more managers in that asset class that are slaying their benchmark than in any other. Yet even from this perspective, it is the overwhelming consensus of investment professionals that a minimum risk portfolio must contain some percentage of equities to meet its liabilities.
Pretenders to the throne
Caught between the need for increased return and a fear of equities, some investors are seeking out asset classes that can act as substitutes to equities and constitute the higher return element of portfolios. The most favoured assets for this role are inflation-linked bonds and hedge funds. On the surface, this seems like a logical move. Both inflation-linked bonds and hedge funds (although it is misleading to group the diverse range of hedge fund strategies under one banner) can provide returns that capitalise on inflationary gains.
Nonetheless, we believe the individual characteristics and risk/reward profiles of these asset classes make their use as like-for-like substitutes to listed equities ill-conceived. To start with, both of these assets have capacity constraints – something not often associated with equity investing. Not all governments offer inflation-linked bonds and those that do have not historically provided enough quantity to satisfy the existing demand, let alone provide a home to the vast pool of assets currently invested in the stock market. On the hedge fund side, there were approximately 5,379 hedge fund managers worldwide at the close of 2002, managing $633bn (e552bn)2. While this may provide enough room for diversification against individual manager risk for private investors, this would pose significant capacity constraints if trying to accommodate the colossal global pool of institutional money.
The return profiles of these two ‘substitute’ asset classes are also very different. Inflation-linked bonds offer a guaranteed minimum return, as well as benefits to increases in inflation, but pricing of these securities has historically resulted in poor returns; they have underperformed nominal bonds over certain time periods. Although hedge funds offer more equity-like returns (potential for significant upside/loss of principal on the downside), there are a number of issues with their performance record, not least the inconsistency of available data. The MSCI Hedge Fund Index shows hedge funds have outperformed US equities since mid-1995, whereas InvestHedge shows outperformance since only the second quarter of 2001, suggesting a reliance on a bear market to generate superior returns; these results are also subject to a survivor bias. Standard deviation of returns between managers is also extremely high when compared to most other asset classes, and small hedge fund teams pose significant organisational risk.
But it is not our intention to rubbish other asset classes. To achieve performance goals, pension fund managers should use a wide variety of asset types when constructing their portfolios. Asset classes are neither good nor bad. They are what they are. You just have to keep them in balance. The point here is that inflation-linked bonds and hedge funds are not simple like-for-like substitutes for equities.
Lessons from the past
If you believe that equities are still a valid component of your portfolio, your exposure to US equities is no doubt significant. In the final section of this article, we identify some of the ways in which US equity portfolios became imbalanced during the last bull run and exposed the diversification myth inherent in indexation.
Let’s take some lessons from the past. Investment in equities must be diversified across size and style. In our experience, with our conservative and measured approach, we have dampened out volatility and done fairly well. Just as we never overindulged in the dot-com frenzy during the bubble, so now we do not shy away from equities, including growth stocks and technology.
Going forward, we continue to believe that stocks that demonstrate high-quality earnings are an important part of a diversified investment strategy. We remain committed to the idea that active management between styles and size can produce incremental outperformance while marinating diversification and matching our clients upcoming liabilities.
Many of the European clients and prospects who we work with have come to T. Rowe Price looking for a broader exposure to the US equity markets. Many of these prospects and clients utilise the MSCI USA Index or S&P 500 as a primary investment benchmark, and have expressed an interest in including growth and value style tilts in their investment programmes, as well as investments in small-cap equities.
As investors, we feel it is important to understand not only what is in a benchmark, but what is not included. The, S&P 500, like the MSCI USA Index, for example, is a fairly focused and primarily large-cap benchmark. Benchmarking to the MSCI USA Index overlooks a significant segment of the US market: small-cap stocks.
Small-cap stocks routinely undergo extended periods of over and underperformance relative to large-cap stocks, and this was particularly evident during the US market downdraft of the last several years. Small-cap stocks were overshadowed by large-cap stocks in the rally of the late 1990s, and this contributed to significant valuation differences between the two sectors. This trend began to reverse itself in early 2000. Looking at the period from January 2000–June 2002, small-cap stocks, as measured by the MSCI US Small-Cap 1750 Index, returned 5.59%, outperforming the –33.85% cumulative return of the MSCI US Index by 39.44% over the 30-month period. While this is an extreme example, it emphasises the importance of maintaining a diversified portfolio, and understanding the intended as well as unintended bets within a portfolio.
Similarly, growth and value styles of investing witness similar cycles of over and underperformance. Growth stocks, particularly large-cap growth stocks had outperformed value stocks in the bull market of the late 1990s. Looking at the growth and value components of the MSCI Prime Market 750 Index over the same 30-month period beginning January 2000, the MSCI Prime Market Value Index returned –1.13%, relative to the –55.59% cumulative return for the Prime Market Growth Index. In more recent months, this trend has been slowly reversing itself as the Growth Index has begun outpacing the Value Index.
Measured exposure to small-cap stocks, combined with strategic over and underweightings to portfolio sectors such as growth and value provide opportunities to enhance portfolio performance over the long run. When structuring portfolios for our European clients, we have sought to replicate the broad US market as represented by the Wilshire 5000 or Russell 3000 (at present the constituent data on the new MSCI indices is not available). A typical allocation would be 44% large-cap growth, 44% large-cap value, and 12% in small-cap stocks, generally with ranges of ±5–10% around these neutral weights, depending on client preferences. We arrived at these weightings by comparing the structure of client portfolios to broad market indices using a number of characteristics including detailed analysis based on size, growth, and value metrics. Additionally, we look closely at historical as well as projected tracking error relative to major market indices.
This structure of modest over and underweighting of market sectors provides the opportunity to add a measured amount of value through asset allocation decision, without introducing an excessive amount of risk through added volatility or tracking error. While there may be periods when there are not significant differences in valuations between growth and value or large-cap and small-cap, it is better to be aware of the bets in a portfolio and to try and align these to take advantage of imbalances in market valuation. This was evidenced in the market downturn beginning in 2000.
Our general objective in making changes in our asset allocation portfolios at T Rowe Price is to take profits from sectors that have done well and reinvest in sectors that are believed to be more attractively valued. When we make changes to the portfolios, we are distinctly not trying to time the market, but rather we generally evaluate our decisions over an intermediate period, such as the next six to 18 months.
Based on our current outlook, we feel that pockets of the large-cap growth market remain richly priced, although many sectors are quite reasonable relative to the potential growth rates of the companies. It is our belief that these sectors of large-cap growth are more likely to outperform relative to large-cap value in an environment of a recovering economy. Consequently, within our asset allocation portfolios, we have been gradually shifting from a period of overweighting large-cap value to one where we are now overweighting large-cap growth.
We remain at our neutral weight for small-cap stocks, and continue to believe that the outlook for small-cap stocks remains favourable, particularly under a scenario of economic recovery. We trimmed our allocation to small-cap back to neutral in the fourth quarter of 2002 in light of the fact that small-cap valuations did not appear as dramatically compelling as they were when small-cap stocks began their cycle of relative outperformance.
To conclude, exposure to US equities has long given investors strong returns. But, as with all investments, commitments to this asset class must be diversified by both capitalisation and style. Our approach is that investors should consider equities as a vital part of their portfolios.
Todd Ruppert is president and chief executive officer of T. Rowe Price Global Investment Services and David Jones is associate analyst of T. Rowe Price Global Investment Services. Special thanks to Andrew Marks, Darrell Riley, Christian Elsmark and other colleagues for their comments on early drafts