The last 10 years have been a disaster,” says Andrew Chapman, investment manager at the John Lewis pension fund in the UK. In that time, the MSCI World index showed an average annual return of just 0.1% (in sterling terms). Globally speaking, stock markets are almost back where they were in late 2001. Interest rates have plummeted, and the estimated size of future pension obligations have ballooned as a result; bond yields are worryingly low. Inflation may be picking up.

Past performance, as regulators insist fund managers should tell customers, is no guide to future performance. But it does seem prudent to ask if equities have become a poor investment for long-term institutional investors. What part, if any, should shares play over the next decade? To answer that, we need to appreciate both of the roles that equities can play in a portfolio: for capital appreciation and income.

Chapman says the last ten years have also been an extraordinary period, politically and economically. “You need to understand that the last decade started with the tech bubble high, followed by the crash.” And then, of course, there was another bull market, followed by the global financial crisis.

We also need to understand how macro-economic analysis may trump bottom-up stock picking—much as it did in a previous lost decade, the 1970s, when oil crises, terrorism and stagflation paralysed frightened markets.

“From 1972 to 1982, the macro-environment was much more important than stock fundamentals, with markets not making much progress apart from beneficiaries of inflation such as gold,” says Donald Wilkinson, the president and chief executive of the asset manager Wilkinson O’Grady. Understanding such trends is crucial to determine the right mix of cash and equities.

“From 1982 to 2000, the situation changed,” Wilkinson says. “Whilst macro was still important with, for example, the 1987 market crash, a portfolio of good companies could consistently outperform.” Price-earnings ratios peaked at 20 to 30 times profits.

Since the turn of the century, he says, understanding the macro-environment became critical once more, and will remain so for the next three or four years. “From 2000, the world changed again with a decade to fifteen years of price/earnings multiples contracting,” Wilkinson says. “The reduction will continue until equities trade at traditional secular bear market levels of 10 or less, with dividend yields at 5% to 6%.”

Chapman at John Lewis is encouraged. The global financial crash and rebound has certainly caused pain and uncertainty, but, he says: “The excesses have been driven out of the systems, so the next decade should be much more reasonable.”

Equities may be better positioned as an investment than they might first appear. But the same cannot be said for the bond markets. Thirty-year yields in developed markets are worryingly low at around 4%.

Neil Dwane, European chief investment officer of the asset manager RCM warns higher inflation in the next decade or so is inevitable, and that will erode the value of bond portfolios. He blames an increase in debt thanks to central banks’ quantitative easing measures, as they attempt to support economies by buying government bonds.

“If you look at the historical record since 1800, there have been eight periods of high inflation when financial assets have lost their real value, in contrast to real assets such as equities,” Dwane says.

“I can’t understand why institutions are selling equities and buying bonds at current yields, which imply reinvest rates of just 2.5%. They will not be able to produce much of a pension on that.”

The shift in dividend yields represents a fundamental change in the relationship between bonds and equities. There are implications for liability-driven investment (LDI) for pension funds.

There are large uncertainties in estimating a pension fund’s liabilities, such as when they will mature. A key issue is forecasting the actual future risk-free bond yields that determine the discount rate.

Current government bond yields clearly do not represent an unbiased estimate for future government bond returns. They do not include the effects of quantitative easing, or the artificial demand stimulated by the effect of rigid LDI approaches to matching.

Government bond yields may be useful for accountants, and as a shorthand for estimating pension fund liabilities. But the calculated discounted value of the liabilities is just an estimate. It is not a true economic figure on which allocations of assets should be based.

The reality is that any economic valuation of a pension fund’s liabilities builds in a margin of error.

The size of that margin is critical, since it determines whether expensive approaches to matching liabilities make any sense at all. What is the logic in matching a liability stream that has large uncertainties associated with it with precisely determined set of cashflows from a portfolio of risk-free government bonds with nominal yields below 4%, when an equity portfolio can give essentially real yields that are significantly higher? There may be short-term volatility, but there is also a very high probability of matching inflation over the long term required for pension funds.

Whipsawing share values have certainly caught out investors during the last decade. Huge volatility in market prices has obscured the fundamental reason for holding equities for the long term, namely dividend flows.

Dividends and dividend growth are much more important than the valuation expansions that drove markets during periods such as the bull markets of the late 1990s. And as Dwane at RCM points out, are the most consistent and powerful contributors to equity returns.

Going back to the idea that equities are ultimately a source of dividend streams, and that the reinvestment of dividends historically produced a major component of total equity market returns, can have profound implications.

The question that some fund managers are exploring is whether there is institutional demand for equity strategies that focus on generating larger and more stable dividend flows. Would they be relevant for institutional investors, or be regarded as just a variant of a value investing?

Veritas for example, manages a global equity income fund that capitalises on the fact that as markets mature and investors develop internationally, more and more companies pay dividends, and are increasing the size of those payments.

Do such dividend-yielding funds look like alternatives to bond funds? If managers focus on the stability and sustainability of dividend streams, then that might require them to reappraise their definitions of risk.

This lies at the heart of the debate regarding the use and abuse of equity indices. Modern portfolio theory has been built on the idea that investors choose between returns, and the risk of those returns (expressed by their volatility—the standard deviation).

The idea of efficient markets leads to the idea of the market index, as the most efficient portfolio. Major capitalisation-weighted indices usually represent that market portfolio.

Using global equity indices to benchmark performance makes eminent sense. But their implicit or explicit use as tools for the management of equity risk needs to be handled with more care.

If a portfolio of stocks has a more stable dividend stream than the global index and it has a lower absolute volatility, then intuitively, this seems less risky for a long-term investor. But the tracking error of such a portfolio against a global capitalisation-weighted index could be high.

The key issue with equities is that dividends cannot be treated as exact matches for liability cashflows, because payouts can be volatile. But given uncertainties in both liabilities as well as asset cashflows, how much extra should a pension fund pay for assets to get more certainty in cashflows given the uncertainties in liability estimates? In today’s world, shorter-term considerations of solvency are clearly important together with the strength of the sponsor’s covenant. But the appetite for equity appetite also varies: geography reflects the historical prejudices that have driven asset allocation across the globe.

As RCM’s Dwane points out, the average German pension fund still has only 3% in equities, while its peers in the UK have 50%. But the UK and US’s high equity allocations are likely to shrink as their population ages and people start claiming their pensions. In contrast, the rest of Europe, as well as Asia, are underweight equities.

Dwane sees a structural realignment of investment: Asia and continental Europe will increase equity exposure, probably through buying shares from US and UK institutions as they sell them to pay pensions.

For any individual pension fund, the ideal solution is therefore not straightforward, which is just as well, given the huge number of advisors who rely on this for a living.

The John Lewis pension fund has reduced its share exposure by moving into hedge funds and infrastructure. But it still keeps 50% of its assets in equities.

Chapman, its investment manager, says: “If you give up on equities altogether, you may as well give up on the capitalist system.”