Uncertainties govern any future projection of the equity risk premium, argue Wim Barentsen and Svetlana Rodionova, making scenario analysis essential to APG's asset-liability matching process

In order to assess the risk-reward trade-off facing our pension fund clients, APG Asset Management, the manager of the ABP pension fund, performs scenario analysis. By quantitatively analysing a number of fundamentally different investment regimes, APG provides inputs for the asset-liability study that helps underpin strategic investment portfolios. The possible pay-offs of investing in equities feature prominently in this exercise.

APG first presents a number of notions the firm applies while modelling equity returns. It then briefly presents its three scenarios, and goes on to elaborate on the main drivers of equity (out) performance in each scenario.

A broadly-accepted scientific and operational theory about the proper level of the equity premium is lacking, although many explanations exist as to why the historic premium is what it is. The premium that should compensate investors for the "dark forces of time and ignorance", in Keynes's words, is highly uncertain. Any projection must deal with intrinsic uncertainties, which is why APG opted for economic scenario analysis instead of one fixed-equity premium number.

In practice, the firm assesses the likely excess return on equity risk by framing the fundamental drivers of equity and government bonds returns in a number of logically consistent, alternative projections. For equities, these include dividend yields, profit margins, sales growth and valuations. Government fixed-income returns are driven by yield levels and their changes, policy rates, and term and inflation-risk premiums. Quantification is supported by small-scale econometric models that explain valuations, profitability and risk premiums in terms of macrofundamentals and risk factors.

Long-term consistency between required and expected returns is assured by imposing accounting-type conditions. One, for example, relates to the split in return between safe and risky assets. This split is driven by the risk-aversion among investors and the prevailing risks in the economy. The other side of this return equation consists of the sources of return being, in the long term, only dividend income (yield) and growth of earnings per share.

The scenario analysis allows for medium-term developments that substantially deviate from the long-term picture. With, among other things, valuations moving from current levels towards those consistent with the structural features of the modelled regimes, realised equity risk premiums are bound to deviate from what is sustainable in the long term.

APG's three scenarios address the medium and long-term uncertainties caused by the current crisis and the (likely future) policy responses:

Watchdogs: the economy improves by fits and starts, while governments and supervisors effectively re-regulate the financial system, making it less pro-cyclical. The G20 and IMF gradually provide for more effective governance of the global economy. The conducive regulation enhances the economy's ability to reach its potential over the cycle.  Rollercoaster: animal spirits revive and the economy recovers strongly on the back of booming asset markets. The pressure for a fundamental overhaul slowly wanes. With hindsight, the financial crisis is re-interpreted as a major incident (an ‘event caused by fools'). On the face of it everything looks fine - until the next crisis. Blocs: it is, ultimately, beyond the reach of policymakers to prevent a protracted period of stagnation. Government initiatives and pressure groups dominate. Tensions are rife, fear and risk aversion re-emerge. Due to several years of economic hardship, many turn inward and a backlash against the global marketplace unfolds.

Profitability in the scenarios is driven by its historic dependence on the level of growth, corporate sector pricing power and labour market tightness. ‘Rollercoaster' features bouts of high profitability, giving way to strong margin compressions. In ‘Watchdogs', profitability is more stable around historic averages, whereas ‘Blocs' features depressed margins. The price-to-earnings (P/E) ratios in the scenarios follow mainly from their historic dependence on the prevailing levels of inflation and growth and the degree of uncertainty of economic environment. Very low and high inflation seriously depress valuations, as does very weak growth and elevated volatility in the real economy. The P/Es in ‘Rollercoaster' exhibit a volatile path. In ‘Watchdogs', valuations converge to the level close to historic average. ‘Blocs' is characterised by depressed levels over a protracted period of time.

Taking this summer's situation as a starting point and translating the scenarios into paths for the return on a global equity portfolio, APG arrived at a range for the estimated long-term excess returns versus government bonds of about 2-4.5%. In ‘Watchdogs', increased regulations result in smaller swings in prices as the regulator tries to limit financial and economic risks. In ‘Rollercoaster', equity returns can be seriously hurt through spoilage of capital in the build-up of booms. Low cost of capital then translates into excess capacity, causing low profits per unit of capital. This, in turn, seriously depresses equity returns. The dismal medium-term developments in ‘Blocs' result in negative excess returns. However, as the higher-risk environment gets priced in, the return potential of equities improves, so that, even in this scenario, equities can outperform bonds.

Unless this financial crisis turns into a period of bad deflation or a protracted period of high inflation, current equity valuations and bond yields add significantly to the chances of equities outperforming bonds over the next decade.

Wim Barentsen is chief strategist and Svetlana Rodionova is senior investment strategist at APG Asset Management