Antti Ilmanen

Expected Returns on Stocks and Bonds

There is an ongoing shift in opinion about expected asset returns. Ex ante equity returns are less transparent and thus more open to question than ex ante bond returns. Long-term equity returns have traditionally been predicted from historical average market returns, but today they are increasingly predicted using dividend discount models. Realized average returns are misleading guides for the future when expected returns vary over time, and our recent bear market has made investors more aware of forward-looking expected return measures; the starting price matters. Neither of two types of “expected returns”—objectively feasible long-run returns and subjective (often hoped-for, extrapolative) return expectations—can be directly observed, but we can try to estimate them by analyzing historical returns, investor surveys, and market valuation indicators. Objectively feasible equity returns have been low for several years (5%–8%), but subjective return expectations have begun to moderate only recently. The divergence between objective and hoped-for returns was extraordinarily wide around 2000; while the gap has narrowed since, it may not have fully closed.

The Death of Diversification Has Been GreatlyExaggerated

Diversification is famously referred to as the only “free lunch” in investing, but it has been under assault since the 2007–2009 global financial crisis, when virtually all longonly asset classes moved down together. Ilmanen and Kizer argue that the attacks are undeserved. Most investors were never as diversified as they thought they were, and there is ample room for improvement by shifting the focus from asset class diversification to factor diversification. They show that diversification into and across factors has been much more effective in reducing portfolio volatility and market directionality than asset class diversification. The benefits are greatest for long–short investing, which requires shorting and leverage but are also meaningful in a long-only context.

The Norway Model

The Norwegian Government Pension Fund Global was recently ranked the largest fund on the planet. It is also highly rated for its professional, low-cost, transparent, and socially responsible approach to asset management. Investment professionals increasingly refer to Norway as a model for managing financial assets. We present and evaluate the strategies followed by the Fund, review long-term performance, and describe how it responded to the financial crisis. We conclude with some lessons that investors can draw from Norway’s approach to asset management, contrasting the Norway Model with the Yale Model.

Which Risks Have Been Best Rewarded

An empirical study examines the consistency of rewards for bearing various types of risks in U.S. asset markets between 1985 and early 2002. Bearing duration risk and equity market risk was amply rewarded, while bearing long-dated credit risk buying credit-risky bonds versus governments realized puny average profits. Bearing short-dated credit risk gave the most consistent profits of all static overweight strategies, with an information ratio near one. Wide break-even spread cushions may be one explanation for the superior reward for risk at short maturities. More fundamentally, market segmentation seems to be the main explanation, because the most consistent profit opportunities involve switching from Treasuries to top-rated non-government bonds.

Asset Allocation and Bad Habits

This article documents the “bad habits” of investors in asset allocation practices. Whereas financial markets exhibit momentum over multi-month horizons but more reversion to the mean over multi-year horizons, many investors act like momentum investors even at these longer horizons. Both these patterns are well known anecdotally but have not been well documented statistically, especially together. This article therefore addresses two empirical questions. First, How do funds reallocate based on past returns? The authors provide direct evidence using the CEM Benchmarking data on pension fund target allocations over a 22-year period. Second, What are momentum/reversal patterns in financial markets returns? Evidence is provided using more than a century of data. Merging the findings from the two data sets provides evidence consistent with the premise that investors chase returns over multi-year horizons, which is likely to hurt their long-run performance. However, the statistical evidence on pro-cyclical multi-year asset allocations and multi-year mean reversion patterns in asset-class returns is on the borderline of statistical significance.

Exploring Macroeconomic Sensitivities: How Investments Respond to Different Economic Environments

A growing number of investors view their portfolios as a collection of exposures to risk factors. Risk-based investing can mean different things to different investors, but the common feature is the emphasis on improved risk diversification. Many investors identify risks primarily as asset class exposures; others may look at underlying macroeconomic exposures, such as inflation sensitivity. The difficulty with the latter approach is that macroeconomic factors are not directly investable. In this article, the authors study the sensitivity of traditional asset classes and dynamic strategies to different macroeconomic environments: growth, inflation, real yields, volatility, and illiquidity. They identify environments that are particularly challenging for investors and find evidence that dynamic systematic strategies, known as style premia, have meaningfully less macro exposure than do asset classes. They also show how diversification reduces portfolios’ macro risk exposures.

Contrarian Factor Timing is Deceptively Difficult

The increasing popularity of factor investing has led to valuation concerns among some contrarian-minded investors, and fears of imminent mean-reversion and underperformance. The authors find that despite their recent popularity, the most common factors or styles are not, in general, markedly overvalued as measured by their value spreads. More broadly, tactical timing, whether of markets or factors, always seems to hold appeal for many. The authors look at the general efficacy of value spreads in predicting future returns to styles. At first glance, valuation-based timing of styles appears promising, which is not surprising because it is a simple consequence of the efficacy of the value strategy itself. Yet when the authors implement value timing in a multi-style framework that already includes the value style, they find somewhat disappointing results. Because value timing of factors is correlated to the standard value factor, it adds further value exposure—but does so intermittently and suboptimally compared to an explicit risk-targeted strategic allocation to value. Thus, tactical value timing can reduce diversification and detract from the performance of a multi-style strategy that already includes value. Finally, the authors explore whether value timing works better at longer holding periods or at extremes, still finding fairly weak results. They find that contrarian value timing of factors is, generally, a weak addition for long-term investors holding well-diversified factors including value, and specifically, it does not send a strong signal on stretched valuations today.

Bad Habits and Good Practices

Good investing results require both good investments and good investors. This article focuses on the latter, addressing some of the habits that may hinder long-term investment performance: multi-year return chasing, under-diversification, and comfort seeking. The authors also offer some insights that may help mitigate some of these pitfalls in moving toward improved investment outcomes.

Balancing on the Life Cycle: Target-Date Funds Need Better Diversification

Traditional life–cycle strategies have some serious shortcomings. In this article, the authors focus on five of those weaknesses: home bias, inflation sensitivity, concentrated risk, sensitivity to episodes of equity market turbulence, and a lack of highly diversifying alternative strategies. Some of these shortcomings, such as home bias, are easier to fix than others, which may require getting comfortable with financial tools like leverage. Using over 100 years of data, the authors show that embracing financial innovation and addressing these shortcomings could have meaningfully improved historical performance. As life-cycle funds become a bigger part of the investment landscape, the authors argue that it’s becoming increasingly important to utilize more modern investment techniques—many of which have long been part of the institutional investing arsenal.

Pronounced Momentum Patterns Ahead of Major Events

Many financial asset measures exhibit a weak continuation tendency. We show that this tendency is much more pronounced in the run-up to major events such as key macroeconomic announcements and central bank meetings. For example, the likelihood of recent yield trends continuing during the week of the U.S. payroll report release is 60% for ten-year Treasuries, compared to the 53% normal likelihood of trend continuation. There are similar pronounced continuation patterns for other bond markets and major currencies, as well as some evidence of a post-event reversal tendency. The main explanation for these regularities is behavioral. Investors tend to cut losing positions (but run winning positions) in advance of some major event.