Lipper Beating the Benchmark March2012

8/2/2019 Lipper Beating the Benchmark March2012

1/9

AUTHOR:

ED MOISSON

HEAD OF UK & CROSS-BORDER RESEARCH, LIPPER+44 20 7542 3218

[email protected]

BEATING THE BENCHMARK

MARCH 2012

Please note that the views expressed in thisdocument are intended as non-consultative and do not

constitute legal advice.

New research examining how successul actively managed mutual unds in Europehave been in out-perorming indices over the past twenty years. This reveals thattypically 40% o equity unds out-perorm their benchmarks, although this gurevaries widely over time and or unds investing in dierent regions.

Active und managers ability to out-perorm their benchmarks sits near the hearto any discussion on the relative merits o active versus passive investing. In broadterms the argument against investing in an actively managed und is that one takeson the additional risk that the und will signicantly under-perorm the index, a riskthat is exacerbated over time by the additional costs associated with such a und.

The argument against investing in a passively managed und is that one not onlymisses out on the possibility o superior returns that an active manager can oer,but also that, in principle, one is guaranteed to under-perorm the index.

The success o the argument or active und management can be seen in the size othe industry. Actively managed equity unds in Europe stand at just under 1.5trn,while index trackers have 160bn and ETFs 139bn. In other words, o the equityunds pot, passively managed products make up less than 17%.

Clearly the case or active und management goes hand-in-hand with the caseor prudent und selection. Indeed an industry has grown up trying to deliver thelatter or investors, with proessional und selectors choosing unds to invest in and

packaging this up as a product o itsel: unds o unds. Assets invested in undso unds in Europe stand at around 370bn noticeably greater than the assetsinvested in passively managed unds.

An interesting postscript to this is that unds o unds investing solely in passiveproducts have also begun to develop, with assets standing at 4bn (mostly in theUK). The reasons or such a development will be looked at in uture research.

This report will examine how successul actively managed mutual unds in Europehave been in out-perorming their benchmark indices over the past twenty years.Needless to say this will not end the active versus passive debate, but it shouldmake a useul contribution to better understanding how successul active undmanagers have been in delivering on their objectives.

INTRODUCTION

Thomson Reuters 2012 All Rights Reserved. This document is or inormation purposes only anddoes not constitute investment advice or an oer to sell or the solicitation o an oer to buy anysecurity o any entity in any jurisdiction. No guarantee is made that this inormation is accurateor complete and no warranties are made with regard to the results to be obtained rom its use. Inaddition, Lipper will not be liable or any loss or damage resulting rom inormation obtained romLipper or any o its aliates.


2/9

BEATING THE BENCHMARK MARCH 2012

2

CURRENTLY ACTIVE FUNDS

he most straightorward means to assess actively managed unds success ineating their benchmarks is to look at their latest perormance gures. To this

nd all actively managed equity and bond unds perormance relative to theirenchmarks was assessed over 1, 3 and 10 years to the end o December 2011.

or equities, the proportion o unds that out-perormed varied rom 26.7% in 2011,0.0% over 3 years and 34.9% over the past 10 years. While bond unds aredetter over 3 years (45.4% out-perormed), the proportion tailed o dramaticallyver the 10-year period, alling to 16.2%. The latest year was similar to that orquities, with 23.7% o bond unds out-perorming their benchmarks.

O course these initial gures do not provide insights into any variations betweenunds investing in dierent markets. Figure 1 takes the same approach as above.e. actively managed unds perormance relative to their benchmark indices over

3 and 10 years) but groups unds by six o the largest equity sectors (dened byipper Global Classications).

unds investing in North American equities have consistently ewer managershat have out-perormed their benchmarks than or other classications. Anotherattern that is consistent across the three time periods is that more unds investing

n UK and European stocks have beaten their benchmarks than those undsnvesting globally. Apart rom these two points, there is little consistency betweenhe three time periods: dierent products out-perorm more commonly overierent periods.

While the primary ocus o the current research is equities, similar analysis was also

arried out or , US$, Global and Emerging Market bond unds. Each classicationroadly ollows the pattern established above or bond unds as a whole. Again themost striking nding is the low proportion o unds that have out-perormed their

enchmarks over a 10-year period. O the our classications, the highest was the3.1% o global emerging market unds that out-perormed their benchmarks. Thether three classications were around 10%, with bonds the lowest and just 8.3%ut-perormed.

he gures or these three bond classications were all lower over 10 years thanver both 1- and 3-year periods. For 3-year returns, the proportions exceeded thesset class total o 45.4%, with the highest being global bonds, where 55.7% o the

unds out-perormed their benchmarks.

REUTERS/HOWARD BURDIT


3/9


3

0%

10%

20%

30%

40%

50%

60%

Dec10-Dec11 Dec08-Dec11 Dec01-Dec11

Eq Asia ex Jap Eq Emg Mkts Equity Europe Equity Global Eq Nth Am Equity UK

-70%

-50%

-30%

-10%

10%

30%

50%

70%

Dec

91-Dec92

Dec

92-Dec93

De-

93-Dec94

Dec

94-Dec95

Dec

95-Dec96

Dec

96-Dec97

Dec

97-Dec98

Dec

98-Dec99

Dec99-Dec00

Dec

00-Dec01

Dec

01-Dec02

Dec02-Dec03

Dec03-Dec04

Dec04-Dec05

Dec05-Dec06

Dec

06-Dec07

Dec

07-Dec08

Dec08-Dec09

Dec

09-Dec10

De

c10-Dec11

-40

-30

-20

-10

0

10

20

30

40

% Equity funds outper for ming benchmar ks M SCI Wor ld (Total R etur n) index

FIGURE 1

FIGURE 2

PROPORTION OF EQUITY FUNDS OUTPERFORMING BENCHMARKS, BY SECTOR

ANNUAL PROPORTION OF EQUITY FUNDS OUTPERFORMING BENCHMARKS


4/9


4

SURVIVORSHIP BIAS AND

ROLLING RETURNS

O course these initial ndings cannot help but partly refect much o the recentmarket turmoil and so a longer term view (beyond the most recent 10 years)

would be invaluable o itsel. More than this, the issue o survivorship bias muste grasped, i.e. average out-perormance gures might be infated - or defated -ue to unds being closed mid-way through a time period. To do this, the currentnalysis looks in greater detail at rolling periods over the past 20 years and

ncludes unds that have since closed or been merged.

Having said this, investors in Europe cannot escape the act that und closuresnd launches are a signicant eature o the industry landscape and that this may

mpact on the unds in which they choose to invest. It is thereore useul or a long-erm investor to at least have a sense that the unds available change each andvery year.

n Europe over the past 10 years, 3,400 unds have been launched every year, onverage, and 2,400 have been closed or merged. In other words, there has been aet increase o 1,000 unds each year (across all asset classes and or both activelynd passively managed unds). The current universe stands at around 35,000

unds, so there is certainly not a shortage o products to choose rom!

unds rolling returns have been assessed over 1, 3 and 10-year periods every yearrom 1992 to the end o 2011. The rst o these, as the shortest period, is mostseul in tackling survivorship bias and gure 2 refects the number o equity

unds that out-perormed their benchmarks each year over this 20-year period.The MSCI World index has been included to provide broad context in the chart,lthough obviously only a relatively small proportion o the unds assessed will beenchmarked against this index).

he proportion o equity unds that have out-perormed their benchmarks hasaried between 59.1% and 26.7%, coincidentally the rst and last years in thisnalysis. The annual average is 42.8%. This last gure is at the higher end o thepectrum ound in the initial analysis presented above, suggesting that the dicultecent market conditions have indeed had a negative impact on the proportion octive managers that have been able to out-perorm their benchmarks.

he 26.7% o actively managed equity unds that out-perormed their benchmarksn 2011 can be urther scrutinised by seeing how many also out-perormed in010. The number o unds achieving this alls by 40%, with 16.4% o unds out-erorming in both years. This process can be extended urther, with 8.6% out-erorming in 2011 and 2010 and 2009; 5.2% out-perorming in each year backhrough 2008; 2.6% through 2007; 1.6% through 2006; 1.3% through 2005; 1.0%hrough 2004; 0.8% through 2003. Finally, 0.5% (13 unds) have out-perormed inach o the ten years rom 2002 to 2011.

REUTERS/CARLOS BARRIA


5/9


5

LONGER-TERM PERIODS

For long-term investors the act that an active manager does not out-perorm inevery calendar year is likely to be less signicant than whether he/she can out-perorm over a longer time period. To examine this, the data has been expanded tolook at rolling 3-year and 10-year periods.

For all three sets o data the rolling period used is 1 year. For example, or 3-yearperormance the periods begin rom December 1991 to December 1994, thenDecember 1992 to December 1995, then December 1993 to December 1996, and soon. This means that or 3- and 10-year rolling returns, a proportion o many undsperormance will be accounted or in more than one period. This helps to accountor any survivorship bias.

The average proportion o 42.8% o equity unds out-perorming their benchmarksover the past 20 years is a gure reerred to above. Figure 3 now puts this into awider context. The equivalent gure or 3-year rolling periods is 41.4% and or10-year rolling periods it is 39.7%. In other words, the proportion o unds out-

perorming over longer periods may have dropped very slightly, but it remainslargely stable. For 10-year returns in particular it is unavoidable that there will havebeen unds that closed beore they reached their tenth year, and more likely theseunds will have been perorming less well.

A more noticeable decline as the periods lengthen can be seen when one turns toxed income. The gure or the asset class as a whole is consistently lower than orequity unds and drops to just 17.4% over rolling 10-year periods.

This latter nding bears out that it is also crucial to drill down and identiy anyvariations or unds investing in dierent regions. For bond unds one can clearlysee that or emerging market and global products or 1- and 3-year rolling periods

the proportion o unds out-perorming is airly similar to that or equity unds. Butor -denominated and US$-denominated bond unds, the proportion o unds isnot only smaller than equities or 1-year returns, but it then declines as the periodsassessed are lengthened, so that just over 6% o unds in these two classicationshave out-perormed their benchmarks when looking at 10-year rolling returns.

On the equity side, while unds investing in Asia and the UK maintain or improveon the proportion o unds out-perorming as the time periods lengthen, theopposite is the case or the our other classications assessed. These two divergingtrends mean that or 10-year periods, 20.8% o North American equity unds out-perormed their benchmarks on average, while the equivalent proportion or AsiaPacic (ex-Japan) unds is 54.4%.

REUTERS/DARREN WHIT


6/9


6

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Dec91-Dec92

Dec92-Dec93

De-93-Dec94

Dec94-Dec95

Dec95-Dec96

Dec96-Dec97

Dec97-Dec98

Dec98-Dec99

Dec99-Dec00

Dec00-Dec01

Dec01-Dec02

Dec02-Dec03

Dec03-Dec04

Dec04-Dec05

Dec05-Dec06

Dec06-Dec07

Dec07-Dec08

Dec08-Dec09

Dec09-Dec10

Dec10-Dec11

Eq Asia ex-Jap. Equity Europe Equity Global Eq Nth Am Equity UK

1 YEAR

PERIODS

3 YEAR

PERIODS

10 YEAR

PERIODS

ALL EQUITY FUNDS 42.8% 41.4% 39.7%

ALL BOND FUNDS 31.6% 24.7% 17.4%

EQUITY ASIA PAC EX-JAPAN 48.4% 48.9% 54.4%

EQUITY EMG MKTS GLOBAL 38.5% 31.1% 24.6%

EQUITY EUROPE 37.7% 35.9% 27.0%

EQUITY GLOBAL 42.2% 38.4% 32.5%

EQUITY NORTH AMERICA 36.2% 30.3% 20.8%

EQUITY UK 46.4% 47.6% 47.4%

BOND EMG MKTS GLOBAL 45.8% 45.4% --

BOND EUR 25.8% 18.5% 6.3%

BOND GLOBAL 34.4% 30.4% 23.1%

BOND USD 25.9% 16.8% 6.4%

FIGURE 3

FIGURE 4

AVERAGE PROPORTION OF FUNDS ROLLING RETURNS

OUTPERFORMING BENCHMARKS, 31/12/1991 TO 31/12/2011

PROPORTION OF EQUITY FUNDS OUTPERFORMING BENCHMARKS

ROLLING 1 YEAR PERIODS


7/9


7

CLASSIFICATION DIFFERENCES

Looking at 1-year rolling periods by classication o equity unds (using LipperGlobal Classications, gure 4), the proportion o unds out-perorming theirrespective indices move broadly but not always in the same direction in anygiven year. Secondly, the range o proportions o out-perormance have generally

narrowed in the most recent 10 years (rom 2002 to the end o 2011) compared tothe earlier decade (1992 to the end o 2001).

Having said this, some und classications have stayed in a narrower range o out-perormance proportions than others. Picking up on unds investing in Asia andthe UK once more, while the ormer has ranged rom 8.3% (2004) to 83.8% (1999)o unds out-perorming their benchmarks, or the latter the range has been muchnarrower, ranging rom 23.1% (2011) to 64.5% (2000).

The narrowing o the range o unds out-perorming between the rst and second10-year periods is even more pronounced when one looks at 3-year rolling returns(gure 5). On the downside this means that no classication has a proportion o

unds out-perorming in excess o 60% that could be seen in the early 1990s orlate 1990s. But on the upside, this means that classications with a proportion oout-perorming unds below 30% has virtually disappeared in the three most recentrolling periods.

Figure 5 also draws out the act that a greater proportion o UK equity managersgenerally outperorm than or other classications, while US equity managers sitat the oot o the table. While the average proportion o Asia Pacic unds out-perorming is actually higher than that or the UK (see gure 3), this is clearly theresult o results posted over the rst 10 years, while the more recent period hasseen a dramatic change or the worse.

Dierences between equity classications are most clear with the 10-year rollingperiods (gure 6), although the broad picture painted above (or 3-year periods)holds true. 10-year periods also reduce the variation in the proportions or eachclassication. This is most notable or UK equity unds (ranging rom a lowpoint o 37.6% to a high o 52.9), while yet again there is considerable volatilityor Asian equity unds (ranging rom 31.3% to 66.7%) and, unortunately, on adownward trajectory.

REUTERS/SRDJAN ZIVU


8/9


8

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

D

ec91-Dec94

D

ec92-Dec95

D

ec93-Dec96

D

ec94-Dec97

D

ec95-Dec98

D

ec96-Dec99

Dec97-Dec00

D

ec98-Dec01

Dec99-Dec02

Dec00-Dec03

D

ec01-Dec04

Dec02-Dec05

Dec03-Dec06

Dec04-Dec07

Dec05-Dec08

Dec06-Dec09

D

ec07-Dec10

D

ec08-Dec11


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Dec91-Dec01

Dec92-Dec02

Dec93-Dec03

Dec94-Dec04

Dec95-Dec05

Dec96-Dec06

Dec97-Dec07

Dec98-Dec08

Dec99-Dec09

Dec00-Dec10

Dec01-Dec11


FIGURE 5

FIGURE 6






9/9


9

CONCLUDING REMARKS

This research has revealed that typically 40% o actively managed equity undsout-perorm their benchmark indices. While clearly this gure has varied over time,it not only represents the proportion o equity unds that have out-perormed inthe most recent 3 years, but also approximates the average proportion o equity

unds that have out-perormed or 1-, 3- and 10-year rolling periods over the past20 years.

While those selecting actively managed unds to invest in may well like theirchances o nding the 4 in every 10 unds that can out-perorm an index, thisproportion does vary each year (broadly between 30% and 60% o unds, see gure2), as well as varying or unds investing in dierent regions (which can be between20% and over 50% over dierent periods, see gure 3).

The ndings presented here will clearly not settle the active versus passive debateone way or the other, but they do provide robust statistical research into activelymanaged unds relative perormance across Europe. Such insights can better

inorm this ongoing discussion.

METHODOLOGY

Finally, some comments on the methodology used. The universe o unds analysedincluded all open-ended mutual unds domiciled in Austria, Belgium, Denmark,Finland, France, Germany, Luxembourg, Liechtenstein, Ireland, Italy, Netherlands,Norway, Spain, Sweden, Switzerland and the UK. Only actively managed equityand bond unds were included, so index tracking unds and ETFs were specicallyexcluded, as were absolute return unds and unds with other alternative strategiesthat tend to use cash-type benchmarks. Funds o unds were also excluded to

enable more consistent comparisons. For any und with multiple share classes, onlythe one identied as the primary class was included.

Analysis o a unds out-perormance was measured using its own benchmarkindex, so where this could not be identied the und was excluded rom the study.Perormance was calculated using total returns, with dividends re-invested onex-dividend date, in each unds local currency. Data points including ewer thantwenty unds were excluded. Last, but by no means least, thanks are due to RanVispap-Rich and Noora Vainio or their invaluable help in preparing this research.

A version o this paper will appear in Institutional Investors Guide to European

Indexing (http://www.euromoneyplc.com)

REUTERS/CARLOS BA

Documents

Lipper Beating the Benchmark March2012