Multi-Asset Class Mutual Funds: Can They Time the Market? Evidence from the US, UK and Canada via SSRN

Andrew Clare

City University London – Sir John Cass Business School

Niall O’Sullivan

University College Cork – Department of Economics

Meadhbh Sherman

University College Cork

Steve Thomas

City University London – Sir John Cass Business School

April 2, 2015

Clare, Andrew D., O’Sullivan, Niall , Sherman , Meadhbh and Thomas, Steve H., Multi-Asset Class Mutual Funds: Can They Time the Market? Evidence from the US, UK and Canada, Research in International Business and Finance, (2015 Forthcoming).

Abstract:

The importance of asset allocation decisions in wealth management is well established. However, given its importance it is perhaps surprising that so little attention has been paid to the question of whether professional fund managers are skillful at timing market movement across asset classes over time. The timing literature has tended to concentrate on the timing skill of single asset class funds. Using data on US, UK and Canadian multi-asset class funds, we apply two alternative methodologies to identify the asset class timing abilities of managers. Overall, whether we apply a returns-based method or a holdings-based testing approach, we find evidence of only a tiny minority of funds with asset class timing ability.

Multi-Asset Class Mutual Funds: Can They Time the Market? Evidence from the US, UK and Canada – Introduction

In this paper we examine the asset class timing ability of a large sample of multi-asset class funds in the US, UK and Canada over the period 2000 to 2012. The interest in such funds continues to grow as investors embrace diversification following two particularly bad experiences with equityconcentrated portfolios since 2000 including the technology stock crash around 2001 and latterly the financial crisis from 2008. Furthermore, as more investors must now take responsibility for their own pension savings in the form of defined contribution savings vehicles, multi-asset class funds are seen as an important ingredient in any practical solution. Individual investors could themselves combine a range of single asset class mutual funds that together comprise a multi-asset class holding. However, it is reasonable to assume that in choosing a multi-asset class mutual fund investors want not just the low cost efficient diversification benefits but also the asset allocation skills of the fund manager. That is, the multi-asset class fund investor is also paying for the manager’s ability to time asset class return movements. An important question therefore, largely unanswered, is whether the managers of such funds possess skill in timing the relative movements of asset classes.

Of course the skills of the multi-asset class fund manager will comprise both the selection of strategic long term asset class weights as well as tactical asset class timing and security selection abilities. In the case of most funds it is impossible to know these strategic weights without detailed interrogation of the trustees and their advisers (though see Andonov et al (2012)). The tactical asset allocation contribution is defined as the difference between the strategic weights and realized allocation weights with the asset class timing component being the over or under-weighting of asset classes relative to the long run strategic target weights. Ibbotson and Kaplan (2000) and Andonov et al (2012) is unusual in having access to strategic policy weights for a sample of pension funds.

They find a roughly equal contribution to returns of 25bp pa from each of policy weights, asset class timing and security selection. Also Blake et al (1999) and find that while UK pension funds did not show superior timing ability across asset classes, specialist managers do possess superior security selection skills. However, multi-asset class mutual funds provide a new context to explore market timing skills since their managers are focused on tactical adjustments to maximize their performance, top their league tables, and attract new capital conditional on asset allocations complying with their generic grouping, such as “Conservative”, “Aggressive”, etc. We present results for a large sample of funds with a variety of asset allocation categories and provide fresh insight into tactical asset class timing skills.

A much-investigated question in finance literature is the return performance attribution of strategic and tactical allocation and security selection. A number of researchers have emphasised the contribution of strategic asset allocation decision: Brinson et al (1986) and Brinson et al (1991) both suggest that asset allocation policy explains more than 90 percent of overall performance while more recent research suggests that strategic asset allocation accounts for only up to 50% of fund performance, the rest being attributable to tactical adjustments and security selection, Ibbotson (2010), Xiong et al (2010).

In a further detailed examination of performance attribution, Daniel et al. (1997) examine ‘Characteristic Timing’ (timing ability of different investment styles which determines whether funds can time portfolio weightings on characteristics such as size, book-to-market ratio and momentum) and ‘Characteristic Selectivity’ (whether funds can select stocks which outperform the average stock having the same characteristics). The authors find that while performance is significant, it is no greater than the difference between passive and active fund expenses. This is a vast literature. Our paper focuses on the tactical asset allocation skills of multi-asset class funds, specifically on monthly asset class timing and contributes to the mutual fund timing literature in particular.

Multi-Asset Class Mutual Funds

See full PDF below.