Investing in Systematic Factor Premiums

• DOI: http://doi.org/10.1111/eufm.12081
• Date: 01 March 2016
• Published date: 01 March 2016

Investing in Systematic Factor

Premiums

Kees G. Koedijk

TIAS School for Business and Society, Tilburg University P.O. Box 90153, 5000 LE Tilburg,

The Netherlands; and CEPR

E-mail: K.Koedijk@tias.edu

Alfred M.H. Slager

TIAS School for Business and Society, PO Box 90153, 5000 LE Tilburg, The Netherlands

E-mail: A.Slager@tias.edu

Philip A. Stork

VU University Amsterdam Faculty of Economics and Business Administration, De Boelelaan 1105,

1081 HV Amsterdam, The Netherlands; and Tinbergen Institute Amsterdam

E-mail: p.a.stork@vu.nl

Abstract

In this paper we investigate and evaluatefactor investing in the US and Europe for

equities and bonds. We show that factor-based portfolios generally produce

comparableor better portfolios than market indices.We expand the analysis to other

asset classes and factors, work with other optimisation methods and add a basic

liability structure.The results do not depend on adding other asset classes or on the

removal of a specific factor. Finally, we study the results for a worldwide investor

who invests beyond the US andEurope. Over the longer term and with consistently

applied factor diversification, factor investing appears to be advantageous.

Keywords: portfolio management, factor investing, diversification, optimisation

JEL classification: G11, G12, G23

1. Introduction

Diversification is one of the key fundamental investment concepts, both in theory and in

practice. It even has the status of being the only truly free lunch in the world of

We thank the Editor (John Doukas) and an anonymous referee for excellent comments and

suggestions. We are also grateful to Robeco and especially David Blitz, Joop Huij, Simon

Lansdorp, Roderick Molenaar and Tom Steenkamp for comments, support and data. We

thank the participants of the Robeco seminars in Amsterdam, The Hague and Frankfurt, and

the seminar participants at Erasmus University Rotterdam for helpful suggestions. Any

errors are the responsibility of the authors. Correspondence: Kees G. Koedijk.

European Financial Management, Vol. 22, No. 2, 2016, 193–234

doi: 10.1111/eufm.12081

© 2016 John Wiley & Sons, Ltd.

investment (Ilmanen, 2011). The idea is that by combining investment categories, you

can limit a portfolio’s price volatility without affecting expected returns. However, since

the 2007–2009 financial crisis, when almost all investment categories declined en masse,

this received wisdom is being increasingly criticised. In short, the investments turned out

to be not as diversified as expected. And that begs the question: Is it then possible to

effectively diversify a portfolio? Does this mean a portfolio has to be structured

differently, or even that another type of portfolio altogether needs to be constructed?

Investing using factors is one such approach. Investors and researchers were already

aware of this approach, and have identified many factors over the years. The interest in

factors only got the boost it needed when in 2009 the Norwegian Government Pension

Fund requested researchers Ang, Goetzmann and Schaefer to study the results of the

fund’s active management. The results of active management turn out to be more closely

related to market developments than had previously been thought. The researchers

analysed this effect, ascribing the results for a large part to factors that demonstrate a

more systematic character, rather than to factors of a more transient nature to which

active managers were more likely to allocate. This result triggered other investors to

question whether they themselves were actually aware of the systematic factors behind

their investments, and how a greater understanding of these factors could improve the

investment process and results.

This article reflects that line of research. This paper provides a broad, comprehensive

overview and extension of research into factor investing, touching upon relevant debates

and developments. We also add to the existing empirical results. Considerable focus has

traditionally been on the US data while this paper focuses on the European as well as the

US market data and factors, (re)affirming the relevance of factor investing for the

institutional investors in Europe as well. One of the contributions of this paper is

the continued focus on relevance for institutional investors, which is visible throughout

the paper. Factor investing is extended for bonds and its effects are tested on portfolios

based on both equities and bonds. Additionally, liabilities are added to the analysis,

mimicking the fundamental choices that trustees and investment committees face.

Overall, the picture emerges that factor investing calls for more knowledge and research

into embedding this into the investment process, due to its apparent advantages from a

fiduciary investment perspective.

The sections are structured as follows. Section 2 places the attention on factor

investing in a wider context. The aim of this section is to examine the relevant literature,

and to answer the question as to why this subject is such a live one. Section 3 then

introduces the factors. What is the theoretical support for them, and what are the pros and

cons? Section 4 practically fleshes out the theoretical view of section 3 with historical

data series. We compare in plain terms how a portfolio based on investment categories

performs differently than a portfolio using factors and we assess the robustness of the

results. Various optimisation methods are compared in section 5. Section 6 studies

factor-oriented investing if liabilities are present. Section 7 summarises the key findings.

2. Factor Investing in Perspective

Factor investing is increasingly in the spotlight. The concept is also known as

‘risk-premia investing’, or investing in risk premiums. Other investors call it smart beta.

What all these terms have in common is that investors try to identify sources of return and

then attempt to define how they can capitalise on them effectively. One new aspect of this

© 2016 John Wiley & Sons, Ltd.

194 Kees G. Koedijk, Alfred M.H. Slager and Philip A. Stork

development is that it goes further than active, versus passive, adjustment to benchmark

or the development of an effective investment strategy. So what is it then? It revolves

around an advanced understanding of the key choices faced by investors.

To handle this subject properly it is worth taking a step back and assessing how and

why investing using factors has gained so much traction of late. We can identify three

distinct periods in this process: the foundation of the investment profession, the quest for

diversification and the subsequent emerging role of factor investing.

2.1. Period 1: foundation

We can comfortably place the first period in the founding years of the profession. Over

the course of the 1970s, academic insights gained greater acceptance among investors.

The growing awareness of the increasing complexity and of complicated market

conditions demanded that investing become a serious discipline (for an overview, see

Bernstein, 2007). For example, Markowitz (1952) developed the concept that divides

investing into three elements: return, risk and correlation. There is a connection between

expected returns and risk exposure. Higher returns can only be realised with the

acceptance of greater risk. Total portfolio risk can be reduced by spreading the portfolio

over investments that are not fully correlated. Sharpe further adapted this model for

investment practice, dividing risk into unavoidable risk and avoidable risk. Part of the

risk, the so-called systematic risk, of each individual investment is unavoidable. It is

caused by general economic developments in the market, or the common characteristics

of the investment itself: equity investors run different risks than bond holders. Systematic

risk cannot be limited by diversification, so this part of the portfolio’s risk profile should

be compensated in its expected return requirements. It is a different matter with the other

part of the risk, so-called specific risk, as that can be reduced using diversification.

In 1976, Stephen Ross further expounded systematic risk in his Arbitrage Pricing

Theory (see Ross, 1976). If the investor bases his expected risk on a systematic risk he is

exposed to, and that systematic risk changes due to various economic influences, is it

then not more practical to relate returns directly to those various economic influences? In

short, changes in expectations can be explained by the sensitivity to different economic

factors. But what are these then? Chen et al. (1986) find that macroeconomic factors –

unexpected inflation, economic growth, investor confidence and changes in the yield

curve –have a clear influence on equity returns. In their view, a good factor is one that is

primarily expressed in unexpected movements, has a distinctly non-diversifiable effect,

for which information is rapidly and easily accessible, and for which a solid economic

argument can be put forward. Because this approach offers all sorts of benefits (such as

the supposed linear relationship between factors and returns), for many asset managers it

forms the basis for their portfolio and risk-management systems.

2.2. Period 2: refinement and diversification

Model development goes hand in hand with the growth of financial markets, especially

since the end of the 1980s. For various reasons, at that time equities were becoming

increasingly attractive. However, from the end of the 1990s the rise of the second period

occurs, with diversification playing a key role. Investors look back to the 1990s as the

period of equity investment. At the same time, there was a growing awareness that this

was not sustainable –as with any investment the risk of a bubble increased over time. The

© 2016 John Wiley & Sons, Ltd.

Investing in Systematic Factor Premiums 195