Dances with al­pha

Finweek English Edition - - Property Compass - Deutsche Bank

OUR IN­VEST­MENT in­dus­try is cur­rently char­ac­terised by an ob­ses­sive, com­pul­sive mantra that gen­er­at­ing al­pha is the cen­tral pur­pose of in­vest­ing. Pro­duc­ing al­pha is cer­tainly sen­si­ble and vi­tal to your busi­ness if you’re an as­set man­ager or hedge fund man­ager but is it what the client re­ally needs?

We have spent two years de­bat­ing and soul-search­ing at CIO level with the largest pen­sion funds, sov­er­eign wealth funds, as­set man­agers, hedge funds and con­sul­tants in Europe, Bri­tain and Scan­di­navia. Our in­ten­sive ef­forts and re­search have con­cluded what long-term in­vestors re­ally, re­ally want from their in­vest­ments are three key com­po­nents for in­vest­ment suc­cess, be­ing: • An ex­cess re­turn ( eg above inflation, li­a­bil­i­ties, bench­mark, etc). •A pos­i­tive

af­ter costs. •A re­turn without

“ex­ces­sive risk”. (See il­lus­tra­tion 1).

Note that the word al­pha or skill-based ex­cess re­turn doesn’t fea­ture in the above con­clu­sion. We need to in­ves­ti­gate each of those com­po­nents fur­ther but we must ap­pre­ci­ate that “value for the client” is rep­re­sented by a tri­an­gle of needs, not just al­pha.


Ex­cess re­turns

When gen­er­at­ing an “ex­cess re­turn” our in­dus­try needs to re­alise there are two ways to ac­com­plish that. The first and most ob­vi­ous is through skill-based in­vest­ing (ie, ac­tive in­vest­ing), where we hire ac­tive man­agers to de­liver an ex­cess re­turn (ie, an al­pha re­turn).

The sec­ond and of­ten un­der­ap­pre­ci­ated source of ex­cess re­turns is ex­cess risk. In fact, the very cor­ner­stone of mod­ern fi­nance with its many de­riv­a­tives (eg, CAPM, APT and new-age risk­fac­tor mod­els) all are based on the sin­gle be­lief that the ma­jor­ity of ex­cess re­turns orig­i­nate from those ex­cess risks and not al­pha. Af­ter all, the cor­rect def­i­ni­tion of al­pha is: “The ex­cess, resid­ual re­turn that re­mains af­ter all re­turns from beta risk-pre­mia have been ac­counted for.” (See il­lus­tra­tion 2).

Proper bench­mark­ing

Only once we re­alise the ma­jor­ity of ex­cess re­turns orig­i­nate from ex­cess risks and not al­pha can we even at­tempt to use the word “bench­mark”. That’s of­ten poorly un­der­stood by ac­tive man­agers as well as in­vestors. Al­pha isn’t the re­turn above a bench­mark ‒ it’s the resid­ual re­turn that re­mains af­ter the re­turns from ALL ex­cess risk-pre­mia have been ac­counted for. As a con­se­quence, to mea­sure al­pha we need to know what our be­tas are! In other words, ev­ery ac­tive man­ager needs to have a risk-fac­tor model that he’s manag­ing against, whether he wants one or not.

A well-de­fined risk fac­tor model de­fines ex­plic­itly which be­tas or risk-pre­mia the fund is ex­posed to. Without such a model we can’t even mea­sure or de­fine al­pha. The prob­lem comes in when we ask ac­tive al­pha man­agers to ex­plain their risk-fac­tor model. If they don’t have one, or don’t un­der­stand what you’re talk­ing about, it’s very dif­fi­cult for them to jus­tify or even claim al­pha. Soon, if ac­tive fund man­agers don’t in ad­vance and ex­plic­itly de­fine their risk fac­tors, a risk-fac­tor model will be im­posed on them (see ar­ti­cle in this sup­ple­ment on “Re­turns-based style anal­y­sis”).

Without a prop­erly de­fined risk fac­tor model, al­pha could mean any­thing. It’s sad but true that much ex­cess re­turn we term al­pha is ac­tu­ally not “real” or “pure” al­pha. Our con­cern is that the al­pha in­dus­try is cur­rently suf­fer­ing from an iden­tity cri­sis and that the term “al­pha” is rapidly be­ing lost in trans­la­tion. Too many fickle in­ter­pre­ta­tions of al­pha have cre­ated a se­ri­ous and awk­ward “trans­la­tion fal­lacy” ‒ where what we think al­pha is, or what we want al­pha to be, is mov­ing away from an ob­jec­tive, sci­en­tific and con­sis­tent def­i­ni­tion of al­pha.

Beat­ing bench­marks without skill

Given the above, is it pos­si­ble to beat a com­mon bench­mark without skill? We’ll now demon­strate two ways to achieve that. The first bench­mark we’ll use is a well­known type of strate­gic as­set al­lo­ca­tion or bal­anced fund ap­proach, where the bench­mark con­sists of the re­turn cal­cu­lated from a 60% weight­ing in eq­ui­ties, 30% in bonds and 10% in cash.

Much re­search has been con­ducted into the eq­uity risk pre­mium (ie, the ex­cess re­turn above the risk-free rate we ob­tain by tak­ing on eq­uity risk). If we as­sume the eq­uity risk pre­mium is around 5%/year then the eas­i­est way to out­per­form our strate­gic al­lo­ca­tion (60%/30%/10%) over the medium to long term is to go over­weight eq­ui­ties, be­cause the eq­uity risk pre­mium is higher than the bond risk pre­mium (2% to 3% above cash re­turns).

So sim­ply by tak­ing on a static higher eq­uity ex­po­sure (eg, 75% in eq­ui­ties) we can out­per­form. That’s not a skill-based de­ci­sion but merely an ex­cess re­turn by over­weight­ing the riskier beta or as­set class. The more risky port­fo­lio nat­u­rally should pro­vide us with an ex­cess re­turn over the medium to long term if ba­sic fi­nan­cial the­ory is sound. That’s clearly not an al­pha re­turn but a sim­ple beta re­turn.

Roland Rousseau

Newspapers in English

Newspapers from South Africa

© PressReader. All rights reserved.