2011 09 PerTrac Survey on HF Age and Size Impacts on Performance for S1 2011

8/4/2019 2011 09 PerTrac Survey on HF Age and Size Impacts on Performance for S1 2011

1/32

Impact of Fund Size and Age onHedge Fund PerformanceFifthAnnualUpdatefor2010Performance,Witha2011Review

September2011


2/32

2011PerTrac Page 2

TABLE OF CONTENTS

OVERVIEW___________________________________________________________________________________3

HIGHLIGHTSFORFIRSTHALFOF2011ANDFULLYEAR2010___________________________________________4

COMPARING2010WITH2009___________________________________________________________________5

TRENDSDURING1996TO2010__________________________________________________________________6

METHODOLOGY_______________________________________________________________________________8

I. SizeandPerformance__________________________________________________________________9

II. AgeandPerformance_________________________________________________________________10

III. AFinalCheckonTheDead_____________________________________________________________11

HEDGEFUNDPERFORMANCEBYSIZEOFFUNDIN2010_____________________________________________12

AProFormaViewoftheSizeBasedIndices____________________________________________________18

FinalConclusionsonPerformanceandSizeofFunds_____________________________________________20

HEDGEFUNDPERFORMANCEBYAGEOFFUNDIN2010_____________________________________________21

AProFormaViewoftheAgeBasedIndices____________________________________________________24

FinalConclusionsonPerformanceandAge____________________________________________________26

2011FIRSTHALFREVIEW______________________________________________________________________27

CONCLUSIONS_______________________________________________________________________________30

ABOUT

PERTRAC

_____________________________________________________________________________

31


3/32

2011PerTrac Page 3

OVERVIEW

This study examines the performance of hedge funds by Age and Size. Building on earlier

PerTracstudiesentitled,AnExaminationofthe ImpactofFundSizeandAgeonHedgeFund

Performance, published in 2009 and 2010, this PerTrac study provides the most recent full

year

(2010)

data

on

hedge

fund

performance.

2010

performance

figures

are

compared

to

historicalfiguressince1996todeterminewhetherperformancetrendsbasedonsizeandage

havecontinued.Weconcludethestudywithafirsthalfreviewof2011.

Forpurposesofthisstudy,smallfundsarefundswithAssetsUnderManagement(AUM)ofless

than$100million,midsizefundsarebetween$100and$500million,andlargefundsareover

$500million. Youngfundsarelessthantwoyearsold,midagefundsaretwotofouryearsold

andtenuredfundsareolderthanfouryears.


4/32

2011PerTrac Page 4

HIGHLIGHTS FOR FIRST HALF OF 2011 AND FULL YEAR2010

ByAge

In2010,youngfundsoutperformedmidagefundsandtenuredfunds.Youngfundsreturned

13.25%,

mid

age

funds

12.65%

and

tenured

funds

11.77%.

In2010,tenuredfundscomprisedmorethan50%ofthehedgefunduniverse,whilemidage

fundsandyoungfundsmadeup22.44%and24.43%,respectively.

Theperformanceofbothmidageandtenuredfundsthroughthefirstsixmonthsof2011was

significantlystrongerthanthesametimeperiodin2010.

Thoughyoungfundsareunderperformingin2011relativeto2010,theycontinuetoleadthe

ageindicesinYeartoDate2011returnfigures.

BySize

Small fundsoutperformedmidsizeand large funds in2010, returning13.04%,11.14%,and

10.99%,respectively.Small fundsbeatmidsize funds insevenoutoftwelvemonths,while

outperforminglargefundsineightoutoftwelve.

Smallfunds, in2010,comprised71.39%ofthehedgefunduniverse,midsizefunds21.17%,

andlargefunds7.44%.

Theperformanceofbothsmallandmidsizefundsthroughthefirstsixmonthsof2011was

betterthantheirperformanceforthesameperiod in2010,whiletheperformanceof large

fundsforthesameperiodswasdown.

The average 2010 AUM of small funds was $26,152,437, midsize funds $225,671,876, and

large funds$1,847,867,623.Small fundscomprise1.25%ofthetotalaverageAUM in2010,

midsizefunds10.75%,andlargefunds88.00%.


5/32

2011PerTrac Page 5

COMPARING 2010 WITH 2009

ByAge

Young fundsoutperformedbothmidageand tenured funds,returning25.19% in2009and

13.25%in2010.Tenuredfundsandmidagefundsperformedbetterin2009than2010.

Negativemonthlyreturnsweremoreprevalent in2010 than in2009. In2009,young funds

were negative for one month, midage funds for one month, and tenured funds for three

months.In2010,youngfundswereintheredforthreemonths,midagefundsforthree,and

tenuredfundsforfour.

Tenuredfundspercentshareintheagedbasedhedgefunduniverseincreasedto53.13%of

theuniverse in2010from51.65% in2009. Midagefundsdeclinedto22.44% in2010from

23.75% in 2009, and young funds experienced the smallest change, declining to 24.43% in

2010from24.60%in2009.

BySize

Small funds outperformed both midsize and large funds in 2010 by returning 13.04%, but

midsizefundsoutperformedsmallandlargefundsin2009byreturning22.61%.Largefunds

hadtheworstperformanceforbothyears.

Thenumberofsmallfundscomprisingthehedgefunduniversedeclinedin2010from2009.

Smallfundsdeclinedfrom74.07%oftheuniversein2009to71.39%in2010,whilemidsize

funds increased from 19.71% in 2009 to 21.17% in 2010, and large funds increased from

6.22%in2009to7.44%in2010.

ThetotalaverageAUMforallthreefundsizesincreasedin2010from2009,withlargefunds

experiencing the greatest increase in AUM. The average AUM for small funds rose 3.48%,

midsizefunds2.96%,andlargefunds4.11%.


6/32

2011PerTrac Page 6

TRENDS DURING 1996 TO 2010

ByAge

Young funds have outperformed both midage and tenured funds in 13 out of the last 15

years. In1999,youngfundsfinished0.13%behindthemidagefunds. In2003,youngfunds

finished0.56%behindthetenuredfundsand0.18%behindmidagefunds.

Midagefundshaveoutperformedtenuredfundsin8outof15yearssince1996.

Thebestperformanceyear forallthree fundageswas1999,withyoungreturning34.54%,

midage34.67%,andtenured25.26%.

The worst performance year for all three fund ages, and the only one in which any of the

funds finished the year in negative territory, was 2008. Young funds fared best during the

crisis,declining 11.31%,followedbytenuredfundsat 17.85%andmidagefundsat 19.46%.

The cumulative total return for young funds is 848.03%, midage 462.47%, and tenured

373.32%overthisperiod.

Young funds have produced better returns with less volatility since 1996. The annualized

compoundrateofreturnforyoungfundssince1996is16.18%,midage12.20%,andtenured

10.92%. The annualized standard deviation for young funds since 1996 is 6.37%, midage

7.04%,andtenured6.77%.

BySize

Smallfundsoutperformedmidsizeand largefunds ineveryyearexceptfor2008and2009.

Smallfundsweretheworstperformersin2008andtheycameinsecondtomidsizefundsin

2009,returning21.50%,whilemidsizefundsgained22.61%,andlargefundsgained18.72%.

Midsize fundsoutperformed large funds in10outof15yearssince1996,allbutoneyear

since2002.

Thebestperformanceyearforsmallandmidsizefundswas1999,whensmallfundsreturned

32.18%andmidsize26.54%.Thebestperformanceyearforlargefundswas2009whenthey

returned 18.72%. The worst performance year for all three fund sizes, and the only one in

which anyof the funds finished theyear in negative territory,was2008.Large funds fared

best,

declining

14.10%,

followed

by

mid

size

funds

at

16.04%,

and

small

funds

at

17.03%.

Thecumulativetotalreturnforsmallfundsis576.91%,midsize370.12%,andlarge317.74%

since1996.


7/32

2011PerTrac Page 7

Smallfundshaveproducedbetterreturns,butwithmorevolatility.Theannualizedcompound

rateofreturnforsmallfundssince1996is13.60%,midsizefunds10.87%,andlargefunds

10.00%.Theannualizedstandarddeviationforsmallfundssince1996is6.95%,midsize

5.94%,andlarge5.96%.


8/32

2011PerTrac Page 8

METHODOLOGY

Tocreatethetotaluniverseoffundstobeanalyzedforthecurrentstudy,wefirstmergedfive

leading hedge fund databases into a master database: BarclayHedge, BarclayCTA,

Hedgefund.net,HedgeFundResearchandMorningstarHedge(formerlyAltvest)foratotalof

26,881

fund

entries.

Pooling

these

databases

provides

a

comprehensive

universe

of

alternative

Indices, Fund of Funds, Hedge Funds, and Commodity Trading Advisors (CTA). To have an

accuratecrosscomparison forthisstudyweselectedonly funds investingdirectly.Therefore,

IndicesandFundsofFundsareexcludedinthefinaluniverse.

Fromaninitial26,881funds,11,402wereautomaticallydeduplicatedusingthePerTracIDde

duplication process. This reduced the total number of funds within the master database to

15,479 records.Of these,28.10% (4,349)were removed because they wereeither Indices or

FundsofFunds;theremaining71.90%(11,130)offundspassedasHedgeFundsorCTAs.Finally,

toeliminateanyerroneouscrosscomparisonsduringtheFundSizeandPerformancesectionof

thestudy,fundsdenominatingtheirassetsincurrenciesotherthanU.S.Dollarswereexcluded.

Fromtheuniverseof11,130HedgeFundsandCTAs,64.30%(7,157)reportinUSDollars.These

7,157fundsarethebaselineforthe2010studyaswellasthe2011review.


9/32

2011PerTrac Page 9

I. SizeandPerformance

FortheSizeandPerformanceportionofthestudythekeystatistictoexamineisAUMforeach

of the baseline funds. Our automatic deduplication tool is the proper starting point for

removing performance biases associated with doublecounts. However, manual exception

reportingoftheAUMdataisrequiredgiventhatanumberoffundssimplyreportbyrepeating

thesametotalAUMofthefundregardlessofthefundrecordand irrespectiveofthevarious

fundclasses(A,B,C,etc.),series(I,II,III,etc.),structures(offshore,onshore,USDfeeder,etc.),

orcompanies(LP,Ltd,LLC,etc.).AsubstantialnumberoffundsdonotdistinguishtheirAUMat

theshare level fromtheAUMat theoverall fund level.Therefore,thesedoublecountswere

removedpriortoaggregation.WhysomefundsrefrainfromreportingAUMsatshare levels is

opentospeculation,butonecanimaginethatitismucheasiertocopythesameAUMseveral

times insteadof inputtingseveraluniquefiguresduring informationsubmissiontothirdparty

informationproviders.AsaresultofthisAUMdoublecountdilemma,wemakethefollowing

assumptionforthesizeandreturnportionofthestudy:

Assumption:EqualAUMacrossanindividualfundbothatitsfundsharesandatthefund

levelarecountedonlyonce.

DecemberwaschosenasthebenchmarkAUMmonthlyfigurebecause it isthe lastmonthof

theyear.Ofthe7,157baselinefundsinthestudy,311hadnothingreportedfortheDecember

2010 AUM and 1,309 had a $0 value. Therefore, we excluded these funds as well, which

amountedtoa22.64%reductioninthesizeofthebaselineuniversefrom7,157to5,537funds.

Wethenappliedamanualreviewandcomparedidenticalfundnames,returns,andAUMsinan

efforttoindentifyfurtherduplications.

After manually inspecting the data, 670 funds (12.10%) were identified as duplicates and

removed.4,867fundsintotalareusedatthecoreoftheSizeandPerformanceportionofthe

study.EachcorefundwasthenanindexconstituentbasedonthesizeofitsAUM:

Small: Lessthan$100millionAUM

MidSize: $100millionAUMto$500millionAUM

Large: Morethan$500millionAUM

The

funds

were

reclassified

on

a

monthly

basis

and

placed

into

a

corresponding

size

index

basedontheirthencurrentAUM.Theperformanceaverageswithineachindexwerecalculated

every month and compiled into a final YeartoDate (YTD) average return figure. The YTD

averagereturnfigurecountsthemonthlyreturnwhenitfirstoccurs.


10/32

2011PerTrac Page 10

II. AgeandPerformance

FortheAgeandPerformanceportionwestartedwiththeuniverseof7,157baselinefunds.The

key statistics examined were the monthly returns and start dates. We manually reviewed

exceptionreportstofurtherremoveanyduplicatesoffunds.

In the Age and Performance portion of the study the month of December was chosen to be

consistentwiththeSizeandPerformancesectionofthestudy.

We checked the funds that reported to multiple third party data providers but reported to

thoseprovidersondifferentdates.Bytakingintoaccounttheseerrors,547funds(7.99%)were

furtheridentifiedandremoved.

The data for the Age and Performance portion was then sorted by December 2010 returns.

Funds that had no reported returns for December 2010 were also removed (311 in total),

leaving 6,299 funds at the core of the Age and Performance portion of the study. Each core

fundwasthenassignedtoanindexbasedonitsageofexistence:

Young: Lessthantwoyears

MidAge: Betweentwoandfouryears

Tenured: Morethanfouryears

Thefundswerereclassifiedonamonthlybasisandplacedintoacorrespondingageindexevery

month based on their thencurrent AUM. The performance averages within each index were

calculatedaccordingtothesamemethodasintheSizeandPerformancesectionofthestudy.


11/32

2011PerTrac Page 11

III. AFinalCheckonTheDead

Oncealloftheduplicatefundswereeliminated,wewishedtovalidatethatwehadaccounted

forany fundsthat failedduring2010.Accounting for failedordiscontinued funds isreported

andcapturedbythirdpartydataprovidersingraveyarddatabases.

The funds for the two portions of the study were crosschecked with the funds in the

BarclayHedge, BarclayCTA, and Hedge Fund Research graveyard databases, which hold a

combined20,749fundrecords.

AftereliminatingIndices,FundsofFunds,nonUSDfunds,andfundsthatceasedinyearsother

than2010,theuniverseofdeadfundsfor2010wasnarrowedtoonly1,428.

FortheSizeandPerformanceportionofthestudyonly27additionaldeadfunds(0.55%)were

furtheridentified,andsubsequentlyremovedwhichresultedinatotaluniverseforthisportion

ofthestudyof4,841funds.

FortheAgeandPerformancesectionthenumberofadditionaldeadfunds identifiedthrough

this process was only 28 (0.44%), which were also removed which resulted in a universe

indexedforthispartofthestudyat6,271funds.

Themethodologyusedwithinthestudyappliesaconservativeapproachtotheinitialuniverse

construction by eliminating any funds not reporting returns or AUMs in December 2010. In

addition,duringtheinitialmanualexceptionreviewofthededuplicationprocess,asubstantial

portionofdeadfundswereremoved.

.


12/32

2011PerTrac Page 12

HEDGE FUND PERFORMANCE BY SIZE OF FUND IN 2010

ThenumberoffundsineachsizeindexisanaverageoftheirmonthlytotalsasseeninTable1.

EachmonththenumberoffundsinthesizeindexwasresetbasedonfundsthencurrentAUM

size.

As

anticipated

in

2010

from

previous

PerTrac

findings,

the

small

fund

index

is

comprised

of

thegreatestnumberoffundspermonth,at3,262 funds;themidsize indexfollowswith967

funds and the large index remains the smallest at 340 funds. The data shows that growth

beyond$500millionAUMisdifficulttoachieve,withonly7.44%offundsmeetingorexceeding

thatmark.Thisisindicativeoftwotrends(beyondtheobviousofthosefundswhichneverdid

or no longerdofeelaneedtoreporttothirdpartydatabases):1)managersprefertostayin

the AUM mid zone to maneuver their portfolio; and 2) raising additional capital beyond the

$500millionmarkprovestobechallenging.

Table1:NumberofHedgeFundswithinSizeIndicesin2010

Regardlessoffundsize,themonthlyperformancefiguressince1996seemtosuggestrelative

synchronicity. Figure 1 below shows that, irrespective of size, funds generally appear to be

movinginthesamedirection.

Figure

1:

Monthly

Fund

Performance

by

Fund

Size

Indices

(January

1996

to

December

2010)

January February March Apri l May June July August September October NovemberDecember Averag

Small 3077 3109 3121 3166 3221 3271 3304 3341 3350 3352 3404 3430 3262.1

Medium 924 932 946 957 953 963 952 963 988 996 1004 1028 967.1

Large 314 319 337 334 333 327 331 330 350 354 367 383 339.9

TOTAL 4315 4360 4404 4457 4507 4561 4587 4634 4688 4702 4775 4841 4569.2

NumberofFunds

8.00%

6.00%

4.00%

2.00%

0.00%

2.00%

4.00%

6.00%

8.00%

Jan96

Jun96

Nov96

Apr97

Sep97

Feb

98

Ju

l98

Dec98

May99

Oc

t99

Mar00

Aug00

Jan01

Jun01

Nov01

Apr02

Sep02

Feb

03

Ju

l03

Dec03

May04

Oc

t04

Mar05

Aug05

Jan06

Jun06

Nov06

Apr07

Sep07

Feb

08

Ju

l08

Dec08

May09

Oc

t09

Mar10

Aug10

SmallFundMonthlyIndex MidSizeFundMonthlyIndex LargeFundMonthly Index


13/32

2011PerTrac Page 13

In 2010, for example, the strongest performance month for all three fund indices was

September(smallfundsreturned4.00%,midsizefunds3.25%,andlargefunds3.12%)andthe

weakest performance month for all three fund size indices was May (small funds declined

2.91%, midsize funds 2.83%, and large funds 2.83%). Even in 2009 this same directional

movement occurred, with May proving to be the strongest performance month for all three

fund indices (small funds returned 5.77%, midsize funds 5.38%, and large funds 4.27%) and

February being the worst month for all three fund size indices (small funds declined 1.54%,

midsizefunds 0.62%,andlargefunds 0.07%).

Nevertheless,thereareinstancesofsignificantoutliersbetweenthethreefundsizeindicesand

each has experienced months in which it has outperformed the other two significantly. For

example, small funds outperformed midsize funds in October 1998 by 2.55%; small funds

outperformed largefunds inApril2009by2.87%;midsizefundsoutperformed largefunds in

November1997by2.18%;andlargefundsoutperformedmidsizefundsinSeptember1998by

2.32%. This makes it challenging tojudge performance trends solely on monthly statistics.

However,thesetrendsbecomeclearerwhenaggregatedfrommonthlytoannualfigures.

Figure2belowshowstheaveragemonthlyfundperformancesofthefundsizeindicesrolledup

into their respective YeartoDate (YTD) cumulative returns. From this annual perspective a

strong trend emerges, with small funds generally outperforming midsize and large funds. In

2008, however, small funds were the worst performers and in 2009, small funds finished

second to midsize funds. But in 2010, small funds returned 13.04%, outperforming midsize

funds (11.14%) and large funds (10.99%) and reclaimed the top position they had held onto

from1996to2007.

Figure2:AnnualFundPerformancebyFundSize(January1996toDecember2010)

20.00%

10.00%

0.00%

10.00%

20.00%

30.00%

40.00%

ReturnReturnReturn ReturnReturn ReturnReturn ReturnReturnReturn ReturnReturn ReturnReturn Return

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

SmallFunds MidSizeFunds LargeFunds


14/32

2011PerTrac Page 14

Intermsofwhatperformancetrendexistsbetweenmidsizeandlargefunds,itismoredifficult

togaugesincetherankingsamongthesetwoindiceshavechangedwithgreaterfrequency.By

looking at the yearly returns for each fund size index in Table 2 below, a general tendency

emerges in favorofmidsize fundsoutperforming large fundssince2002,excludingthecrisis

yearof2008.Whilethenumberofyearsinwhichsmallfundshaveoutperformedbothmidsize

andlargefundsislarge(13of15yearssince1996),thenumberofyearsinwhichmidsizefunds

haveoutperformedlargefundsisalsonoteworthy(10of15yearssince1996).

Table2:YearlyReturnsforFundSizeIndices(January1996December2010)

A

quick

way

to

make

an

assessment

of

which

fund

size

indices

have

generally

outperformed

the

others isbyviewingthecumulativereturnfortheentirehistoricalperiod.Figure3showsthe

cumulativehistoricalreturnforeachofthefundsizeindicessinceinception.

Figure3:CumulativeReturnPerformancebyFundSize

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2

Return Return Return Return Return Return Return Return Return Return Return Return Return Return Re

SmallFunds 24.89% 20.15% 8.53% 32.18% 16.40% 11.96% 5.70% 24.70% 12.17% 12.41% 14.01% 11.74% 17.03% 21.50% 13

MidSizeFunds 16.62% 17.17% 5.92% 26.54% 12.85% 7.34% 3.92% 17.13% 9.44% 11.32% 13.24% 10.27% 16.04% 22.61% 11

LargeFunds 18.63% 18.05% 6.72% 18. 50% 12. 37% 7. 69% 3. 68% 15. 46% 7.28% 9. 00% 11.61% 10. 22% 14.10% 18.72% 10


15/32

2011PerTrac Page 15

Historicallyfundswithlessthan$100millionAUMhavehadthelargestcumulativereturn.The

cumulativereturngeneratedbysmallfundsoverthe15yearperiodis576.91%,midsizefunds

370.12%, and large funds 317.74%. However, most investors prefer using the Value Added

Monthly Index (VAMI) as a fund evaluation method. VAMI is a statistic used to track the

monthlyperformanceofahypothetical$1,000withcompounding. AccordingtotheVAMI in

Table3,attheendof2010,ifanindividualorinstitutionhadinvested$1,000ineachfundsize

indexduringthefirstmonthoftheirinceptions,the$1,000inthesmallfundindexwouldhave

grownto$6,769,themidsizefundindex $4,701,andinthelargefundindex$4,177.

Table3:CumulativeReturnandEndingVAMIbyFundSizeIndex

(January1996toDecember2010)

CumulativeROR 576.91% 370.12% 317.74%

EndingVAMI $6,769.10 $4,701.24 $4,177.35

SmallFund

Index

MidSize

FundIndex

LargeFund

Index


16/32

2011PerTrac Page 16

An analysis of returns would be incomplete without an examination of their performance

measures.Table4providestheannualizedperformancemeasuresforallthreefundsizeindices

overthestudyperiod:

Table4:AnnualizedPerformanceMeasuresbyFundSizeIndex

(January1996December2010)

The Compound ROR is the monthly average return required for each period so that when

accumulated itwillmatchthefigureforthefinalcompoundedperformancereturnattheend

ofallperiods. It isusedtocalculatetheVAMIand inthisstudyCompoundROR isannualized

along with all other statistics. The study uses a 5% risk free rate of return in keeping with

previousstudies.

Statistically:

StandardDeviationmeasuresthevolatilityofreturnsfromitsmean;

SemiDeviationmeasuresthevolatilityofreturnsbelowthemean;

Gain Deviation measures the volatility of returns from its mean only during periods of a

gain,andLoss Deviationistheinverse;

Downside Deviation measures the potential loss that may arise from risk as measured

againstaMinimumAcceptableReturn(MAR),whichthistableoffersat10%,5%,and0%;

TheSharpeRatiomeasurestheriskadjustedreturninorderto determinerewardperunit

ofrisk.ThehighertheSharpeRatiothebetteristhehistorical riskadjustedperformance;

TheSortinoRatio isanadjustmentontheSharpeRatiothatmeasuresdownside volatility

asexpressedbysubtractingtheriskfreereturnorMARfromthemeanannual returnofthe

portfolioandthendividingbythedownsidedeviation.

From the figures in Table 4, the small fund index provides the greatest Compound ROR

(13.60%)butatthecostofhavingthehigheststandarddeviation(6.95%).WithaSharpe(5%)

ratioof1.17andaSortino (5%) ratioof1.95, thesmall fundsalso generate themost excess

CompoundROR 13.60% 10.87% 10.00%

StandardDeviation 6.95% 5.94% 5.96%

SemiDeviation 7.57% 6.10% 6.01%

GainDeviation 4.51% 4.10% 4.20%

LossDeviation 4.90% 4.18% 4.08%

DownDev.(10.00%) 4.67% 4.17% 4.25%

DownDev.(5.00%) 4.07% 3.54% 3.58%

DownDev.(0%) 3.52% 2.98% 3.01%

Sharpe(5.00%) 1.17 0.95 0.82

Sortino(10.00%) 0.70 0.19 0.00

Sortino(5.00%) 1.95 1.55 1.31

Sortino(0%) 3.64 3.48 3.18

SmallFund

Index

MidSize

FundIndex

LargeFund

Index


17/32

2011PerTrac Page 17

returns, but they have the highest volatility figures in all deviation categories. Midsize and

large funds, however, have very similar deviation statistics. Midsize funds have a lower

standard deviation (5.94% midsize, 5.96% large) and gain deviation (4.10% midsize, 4.20%

large), while large funds have a lower semi deviation (6.10% midsize, 6.01% large) and loss

deviation (4.18% midsize, 4.08% large). What this implies is that it will come down to a

subjectivejudgmenttodeterminewhichfundsizeindexhasmorevolatility.Nevertheless,itis

indicative that the midsize funds generally have outperformed large funds since their

CompoundRORis0.87%greaterandtheirexcessreturnsbasedonacomparisonbetweentheir

SharpeandSortinoratiosarelarger.

Duringthe last15years,smallfundshavegenerallyoutperformedmidsizeand largefunds in

termsofcumulativereturnandannualizedreturns.Partofthisisduetoscalingasafundwitha

smaller capital pool is more likely to return more over an annual period. This has been

addressed in other studies that use data to prove the widely held belief that better returns

comefromsmallerfunds.1

1SiliconValleyBank,DialingDown,5/13/2010andTheSmallFundAdvantage,LarryChengpostedinGrowth

Equity,VentureCapital5/13/2010


18/32

2011PerTrac Page 18

AProFormaViewoftheSizeBasedIndices

ProFormaAnalysiswasperformedusingMonteCarlotoolstosimulateeach fundsize index.

Thecriteriaincluded:

Historical

Monthly

Data

for

each

fund

size

index

(January

1996

to

December

2010)

Fiveyearforwardtimerange

S&P500TR(TotalReturn)asthemarketbenchmark

Riskfreerateofreturnof5%

MinimumAcceptableReturn(MAR)of5%

BootstrapMethod

10,000simulationswithquarterlyrebalancing

Therationalefor includingthesecriteriaare:1)TheS&P500 index isaproxyforthebroader

U.S. market performance; 2) the bootstrap method, by forecasting beyond one period by

relyingontheforecastingdataforthatperioditself,isastraightforwardtechniqueofassigning

measuresofaccuracytosampleestimates;and3)A5%Rfratewasusedwitha5%MARsothat

resultscanbecomparedtopreviousstudies.TheMonteCarloresults forthesmall,midsize,

andlargefundsizeindicesareinTables5,6,and7,respectively:

Table5:SmallFundIndexMonteCarloSimulation

Table6:MidSizeFundIndexMonteCarloSimulation

SmallFundIndexAnnualized

Return

Maximum

Drawdown

NumberSi mul ati ons 10,000 10,000

Mean 13.64% 6.37%

Median 13.62% 6.21%

StandardDevi ati on 3.53% 2.77%

Maximum 27.94% 22.07%

Minimum 0.96% 0.54%

MidSizeFundIndexAnnualized

Return

Maximum

Drawdown

NumberSi mul ati ons 10,000 10,000Mean 10.89% 5.45%

Median 10.89% 5.41%


Maximum 22.66% 19.90%

Minimum 0.10% 0.61%


19/32

2011PerTrac Page 19

Table7:LargeFundIndexMonteCarloSimulation

By comparing Tables 5, 6, and 7, the small fund index shows the potential for the greatest

volatility with a simulated annualized standard deviation of 3.53%, while the simulated

annualized standard deviation for midsize funds is 2.95% and large funds almost equal at

2.94%.Furthermore,thesmall fund indexalsohasthepotential forthegreatestdecline; the

simulatedmaximumdrawdown for thesmall funds is 22.07%,midsize funds is 19.90%and

largefundsis 19.05%.Yetthesmallfundindexhasabetterpotentialtooutperformthemid

sizeand largeduringthenextfiveyears.Themeansimulatedannualizedreturnforthesmall

fundindexis13.64%,formidsize10.89%,andforlarge10.01%.Figure4belowshowsthefive

year cumulative return potential range for the small fund index between the 95th

percentile

(19.46% annualized return potential) and 5th

percentile (7.85% annualized return potential),

alongwithitsmedianandMAR:

LargeFundIndexAnnualized

Return

Maximum

Drawdown


Mean 10.01%

5.55%Median 10.00% 5.23%


Maximum 23.60% 19.05%

Minimum 0.38% 0.78%

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

140.0%

160.0%

Year 1 Year 2 Year 3 Year 4 Year 5

95th Percent ile - 5th Percent ile Minimum Acceptable Return Median

Figure 4: Five Year Cumulative Return Potential for Small Fund Index


20/32

2011PerTrac Page 20

FinalConclusionsonPerformanceandSizeofFunds

In2010,thesmallfundindexregainedthetitleasthebestperformer.Priorto2008,smallfunds

beatmidsizeandlargefundsonaconsistentbasis.Butin2008theonlynegativeyearforany

fund indexsmall fundsweretheworstperformers,declining 17.03%. In2009,small funds

came

in

second

behind

mid

size

funds

in

performance.

While

small

funds

have

generally

outperformedbothmidsizeandlargefunds,theirriskprofileremainsthehighest.Thestatistics

fromtheMonteCarlosimulationshowthistrendlikelycontinuingintheshortandintermediate

terms.


21/32

2011PerTrac Page 21

HEDGE FUND PERFORMANCE BY AGE OF FUND IN 2010

ThenumberoffundsineachAgeindexisanaverageoftheirmonthlytotals.Eachmonththey

were reset based on their thencurrent age. In 2010, reporting funds tend to be older with

53.13%oftheuniversecomprisedoftenuredfunds,whileyoungfunds(24.43%)andmidage

(22.44%)

represent

a

combined

46.87%.

Table8:NumberofFundswithinAgeIndicesin2010

Age,however, isno indicatorofperformance.Young fundsandmidage fundsoutperformed

tenured funds in 2010, returning13.25%,12.65%,and11.77%, respectively.As isthecase in

theSizeandPerformancesectionofthisstudy,historically,eachindexhashadmonthswhereit

has outperformed the other two. For example, in April 1998 young funds outperformed

tenured funds by 3.49% and in August 1998, tenured funds outperformed midage funds by

2.61%.

Figure5:MonthlyFundPerformancebyFundAgeIndex(January1996toDecember2010)

Since1996,however,andonanaggregateyearlybasis,youngfundshaveoutperformedboth

midageandtenuredfunds in13outof15years.Figure6andTable9showstheannualfund

performancebyage.Youngfundsfinished0.13%shortofthetopperformingmidagefundsin

J anuary February March A pri l May J une J ul y A ugust September O ctober N ovember D ecember A verage

Young 1449 1503 1468 1487 1494 1488 1464 1450 1463 1440 1444 1436 1465.50

MidAge 1358 1282 1336 1339 1327 1336 1356 1341 1343 1378 1385 1376 1346.42

Tenured 2897 2973 3009 3049 3107 3159 3211 3288 3320 3360 3411 3459 3186.92

TOTAL 5704 5758 5813 5875 5928 5983 6031 6079 6126 6178 6240 6271 5998.83

NumberofFunds

10.00%

8.00%

6.00%

4.00%

2.00%

0.00%

2.00%

4.00%

6.00%

8.00%

Jan

96

Jul96

Jan

97

Jul97

Jan

98

Jul98

Jan

99

Jul99

Jan

00

Jul00

Jan

01

Jul01

Jan

02

Jul02

Jan

03

Jul03

Jan

04

Jul04

Jan

05

Jul05

Jan

06

Jul06

Jan

07

Jul07

Jan

08

Jul08

Jan

09

Jul09

Jan

10

Jul10

YoungFundMonthlyIndex MidAgeFundMonthlyIndex TenuredFundMonthly Index


22/32

2011PerTrac Page 22

1999 and finished last in 2003. Interestingly, young funds fared best during the 2008 crisis,

limiting their decline to 11.31%, while midage funds fell 19.46%, and tenured funds fell

17.85%

Figure6:AnnualFundPerformancebyFundAge(January1996toDecember2010)

Table9:YearlyReturnsforFundAgeIndices(January1996December2010)

30.00%

20.00%

10.00%

0.00%

10.00%

20.00%

30.00%

40.00%

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

Retu

rn

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

YoungFunds MidAgeFunds Tenur edFunds

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 201Return Return Return Return Return Return Return Return Return Return Return Return Return Return Retu

YoungFunds 29.14% 24.17% 11.61% 34.54% 20.44% 14.27% 8.63% 22.77% 12.76% 14.10% 15.29% 15.02% 11.31% 25.19% 13.2

MidAgeFunds 22.74% 16.41% 5.83% 34.67% 16.45% 10.64% 4.61% 22.95% 10.94% 10.62% 12.56% 9.45% 19.46% 21.51% 12.6Tenured Funds 18.28% 16.92% 6.60% 25.26% 10.80% 8.72% 2.80% 23.33% 10.35% 10.87% 12.71% 9.53% 17.85% 21.01% 11.7


23/32

2011PerTrac Page 23

Thislimitoflossesbyyoungfundsrelativetomidageandtenuredfundsduringthe2008crisis

along with their outperformance of midage and tenured funds in 2010 has helped them in

achieving the best historical performance statistics. Table 10 below shows the annualized

performancemeasurementsforthethreeageindices:

Table10:AnnualizedPerformancebyFundAge(January1996December2010)

As in PerTracs 2009 study, the young fund index continues to have the highest annualized

compound ROR in 2010 at 16.18% (12.20% for midage and 10.92% for tenured) and it

continuestohavethe lowestannualizedstandarddeviation in2010at6.37%(7.04%formid

ageand6.77%fortenured).Furthermore,theyoungfundindexexhibitsthebestrisk,relative

to return,measuresoutofall threeage indices. Itpresents lower figures foreveryvolatility

categoryexceptforthegaindeviation,whichwasatiewiththetenuredfundindex.TheSharpe

and Sortino ratios for the young fund indexarealsostronger than thoseof the midageand

tenuredindices.

These findingssuggestthatyoung fundsmaybeabletoachievehigherreturnswith lessrisk.

Onaverage87%offundsaresmallfundsaswellasyoungfundsandthereforesubjecttothe

scalingissuesasdiscussedinthesizesectionofthisstudy.Youngfundsmaybenefitfromother

advantages which include their ability to conduct portfolio changes quicker and under the

radar;theirlessmatureadministrativeandoperationalneedsresultinginlowerfixedcosts;and

finally, their quick adoption of new technologies that allow them to perform their activities

moreefficientlyinmorescalableenvironments.

CompoundROR 16.18% 12.20% 10.92%

StandardDeviation 6.37% 7.04% 6.77%

SemiDeviation 6.64% 7.59% 7.17%

GainDeviation 4.44% 4.56% 4.44%

LossDeviation 4.11% 5.44% 4.44%

DownDev.(10.00%) 3.85% 4.96% 4.79%

DownDev.(5.00%) 3.28% 4.36% 4.12%

DownDev.(0%) 2.76% 3.82% 3.51%

Sharpe(5.00%) 1.63 0.98 0.85

Sortino(10.00%) 1.44 0.40 0.18

Sortino(5.00%) 3.11 1.53 1.34

Sortino(0%) 5.47 3.03 2.96

Young

FundIndex

MidAge

FundIndex

Tenured

FundIndex


24/32

2011PerTrac Page 24

AProFormaViewoftheAgeBasedIndices

Using the same parameters that were laid out in the Monte Carlo section of the Size and

Performanceportionofthestudy,simulationswereperformedoneachofthethreeagebased

indices. According to a comparison between Tables 11, 12, and 13, young fundsare likely to

continueprovidingthehighestreturnswiththelowestvolatilityduringthenextfiveyears.

Table11:YoungFundIndexMonteCarloSimulation

Table12:MidAgeFundIndexMonteCarloSimulation

Table13:TenuredFundIndexMonteCarloSimulation

The possible annualized mean return for the young fund index is 16.21%, for midage fund

index is 12.24%, and for tenured fund index is 10.96%; the simulated annualized maximum,


Mean 16.21% 4.78%

Median 16.17% 4.57%


Maximum 29.80% 18.36%

Minimum 4.59% 0.36%

YoungFundIndexAnnualized

Return

Maximum

Drawdown

MidAgeFundIndexAnnualized

Return

Maximum

Drawdown


Mean 12.24% 7.03%

Median 12.20% 7.02%StandardDevi ati on 3.54% 3.11%

Maximum 26.54% 25.03%

Minimum 0.56% 0.70%

TenuredFundIndexAnnualized

Return

Maximum

Drawdown


Mean 10.96% 6.62%Median 10.94% 6.46%


Maximum 24.58% 22.62%

Minimum 1.29% 0.80%


25/32

2011PerTrac Page 25

minimum, and median returns are also greatest for the young fund index. For the simulated

annualizedstandarddeviationstatistic,youngfundscarrythelowestvolatilityofthethreeage

indices, at 3.30% compared to 3.54% for midage funds and 3.37% for tenured funds. And

youngfundsalsohavethesmallestsimulatedmaximumdrawdownat 18.36%comparedto

25.03% for midage funds and 22.62% for tenured funds. In continuity with the last two

PerTrac studies, the midage fund index still displays the highest simulated maximum

drawdown.


26/32

2011PerTrac Page 26

FinalConclusionsonPerformanceandAge

In2010,theyoungfundindexoutperformedboththemidageandtenuredindices.Thismakes

theyoungfundindexthetopperformersevenyearsrunningsinceits2003slideintolastplace.

Yet perhaps the most striking finding within this study is that young funds have generally

outperformed their midage and tenured peers while keeping a lower volatility profile. The

statistics from the Monte Carlo simulation anticipate this trend continuing in the short and

intermediateterms.


27/32

2011PerTrac Page 27

2011 FIRST HALF REVIEW

YoungfundsareleadingtheageindicesinYeartoDate2011returnfigures.Table14showsthe

six month performance figures for the age indices and Figure 7 provides a graphical

representationoftheir2011performanceYTD:

Table14:MonthlyandYTDFiguresforAgeIndices

Figure7:MonthlyPerformanceofAgeIndices(January2011toJune2011)

SixMonthPerformanceforAgeIndices(2010and2011)

2010 January February March April May June YTD

Young 0.29% 1.15% 2.49% 1.41% 2.09% 0.10% 2.53%

MidAge 0.77% 0.73% 2.79% 1.38% 2.93% 0.44% 0.66%

Tenured 0.87% 0.85% 2.92% 1.37% 3.26% 0.92% 0.03%


Young 0.18% 1.34% 0.57% 1.67% 0.81% 1.00% 1.94%

MidAge 0.54% 1.00% 0.28% 1.66% 1.16% 1.22% 1.07%Tenured 0.27% 1.37% 0.11% 1.73% 1.46% 1.34% 0.64%

2.00%

1.50%

1.00%

0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

Jan11 Feb11 Mar11 Apr11 May11 Jun11

Young MidAge Te nur ed


28/32

2011PerTrac Page 28

Size indicesshowthattheperformanceofsmallandmidsize funds inthe firstsixmonthsof

2011 is better than large funds. Table15showsperformance figures for thesize indices and

Figure8providesavisualreferencewithregardstotheir2011performanceYeartoDate:

Table15:MonthlyandYTDFiguresforSizeIndices

Figure8:MonthlyPerformanceofSizeIndices(January2011toJune2011)

SixMonthPerformanceforSizeIndices(2010and2011)


Small 0.89% 1.11% 2.79% 1.45% 2.91% 0.67% 0.78%

MidSize 0.47% 0.72% 2.68% 1.21% 2.83% 0.62% 0.60%

Large 0.05% 0.48% 2.81% 1.33% 2.83% 0.81% 0.94%


Small 0.27% 1.31% 0.23% 1.71% 1.24% 1.23% 1.02%

MidSize 0.52% 1.27% 0.17% 1.61% 1.25% 1.24% 1.05%

Large 0.59% 1.35% 0.16% 1.68% 1.23% 1.36% 0.83%

2.00%

1.50%

1.00%

0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

Jan11 Feb11 Mar11 Apr11 May11 Jun11

Small MidSize Large


29/32

2011PerTrac Page 29

AsofJune2011,youngfundshavecontinuedtheirhistorictrendofoutperformancewithlower

volatility. Young fundscarrythe lowestmonthlyaveragestandarddeviation,at1.09%,while

the monthly average standard deviation for midage funds was 1.16% and for tenured funds

was 1.33%. The cumulative return for young funds was 1.94%, midage funds 1.07%, and

tenured funds 0.64%.Figure 9 below showsan annualizedperformance andvolatility scatter

plot:

Figure9:ScatterPlotbetweenPerformanceandVolatility


30/32

2011PerTrac Page 30

CONCLUSIONS

The2010findingscontinuetosuggestthat investorsseekingtomaximizetheirreturnsshould

examine funds with less than $100 million in AUM and funds with less than two years of

existence. Investors need tomatch their liquidityneedsandotherallocation requirements in

their

search

for

small

and

young

funds.

In

keeping

with

historical

expectations,

small

and

young

fundshadgreater returns than theircompetition in2010.Bothhavegenerallyoutperformed

withintheirrespectivecategories;butwhilesmallfundscarrythehighestvolatilityrelativeto

theircompetitors,youngfundshavethelowestintheircategory.


31/32

2011PerTrac Page 31

ABOUT PERTRAC

PerTracprovidessophisticatedsoftwaresolutionsforinvestmentprofessionals,including

pensions,familyoffices,hedgefunds,longonlymanagers,endowments,sovereignwealth

funds,fundsoffundsandindustryserviceproviders.

Oursolutionsspanthecontinuumoftheinvestingprocessincludingfundselection,portfolio

construction,portfoliomonitoring,performanceanalysis,andriskassessment.Ourproducts

alsoenhancecommunicationandreportingbetweeninvestorsandfundmanagers.

Morethan1,400organizationsin50countriesrelyonPerTracsoftwaresolutionstohelpthem

maximizereturns,reduceriskandoperatemoreefficiently.

Formoreinformationpleasevisitwww.pertrac.com.


32/32

Documents

2011 09 PerTrac Survey on HF Age and Size Impacts on Performance for S1 2011