Macro attribution inputs

Anyobdy else find the definition of the normal portfolio noted in 6.3 page 152 volume 6 constrained:

It says " By “normal” we mean a neural position that the fund sponsor would hold in order to satisfy long term investment objectives and constraints".

If there is a normal portfolio it would only exists at points where all asset prices are at equilibrium. Out of equilibrium, while there would still exist an equilibrium allocation, the allocation a fund would need to hold in “order to satisfy long term investment objectives” may be different to lesser greater extent.

In other words, the normal or neutral policy allocation may not necessarily be the allocation you would hold in “order to satisfy long term investment objectives”: t … The neut. Th: he neutral allocation, if out of equilibrium, may deliver a lower than required rate of return.

The normal portfolio is a reference allocation based on general equlibrium precepts and may or may not be the allocation held to achieve long term objectives, given that an allocation should optimise risk, return and relative price movements, which will differ from the equilibrium allocation whenever you are out of equilibrium. Therefore the definition cannot be applied to real world mechanics.

maybe constrained, or whatever.

why are you seriously try to “counterdict” (i made up that word – instead of contradict) the curriculum now? It is what it is. If it is not right, you need to know the not right aspect if ever it gets asked on the exam, come 90- or so days.

after that, go ahead write your thesis, and get a doctorate.

The foundation of the CFA is its ethical imperative, is it not? This implies questioning that which you feel does not add up. Obviously we are not going to change the 2012 exam, that is not the objective. But, if we do not question then perhaps all we are doing the CFA for is the designation alone, which I would have thought undermines its founding and guiding principles.

I do not know how one can learn (not memorise) an answer without first asking and then answering the question.

they are 1. not talking about normal from an equilibrium point of view ever. but normal considering client’s risk and return objectives and constraints present. Hence your normal portfolio is going to be different from mine - and the asset classes on it are also different - given the SAA arises out of the same parameters above.

Given that the IPS is supposed to hold long term - I see nothing wrong in what they are saying / implying.

when prices change, due to the equilibrium forces -> you would have some amount of rebalancing to go back to the strategic asset allocation -> provided the IPS remained the same.

An MVO derived allocation depends on an equilibrium assumption: that is the investment universe is expected to generate expected mean returns, standard deviations and correlations irrespective of historic price movemenst and it is these inputs MVOs used to construct a portfolio.

The long term equilibrium allocation would not change unless there is a fundamental change to expected values of mean return, std dev and corr, but on an ongoing basis price movements (which are assumed to be independent, and random) will differ from the mean expected values and force allocations to deviate from the policy allocations. If you hold the same long term mean expected inputs then any allocation deviation is either short term tactical (you believe you have superior information about some factor or event) or involuntary due to price movement (corrected by rebalancing). It is these short term deviations which are assessed to see whether they add value relative to the policy allocations.

The trouble is that once you move out of a framework that does not assume general equilibrium, independent and random price movement, expected mean returns, standard deviations and correlations change. Therefore there is no static rigid policy allocation, there is no long term static IPS asset allocation.

You can only rebalance back to strategic allocation if the strategic allocation is static, and a strategic allocation can only be static if expected mean returns, standard deviations and correlations remain the same: in other words, it does not matter what the historic price movements were, equilibrium resets and expected price movements remain the same. Only a change in the nature of equilibrium will force a change in returns and price sensitivities.

Therefore, out of equilibrium you have 3 allocations - the equilibrium allocation, the out of equilibrium allocation which benchmarks to the equilibrium allocation, and the deviation from the out of equilibrium allocation. Each personal allocation would also, as you say, adjust for risk aversion and liability profile/ability to bear risk. The allocation universe is 3 dimensional as opposed to the 2 dimensional general equilibrium universe of the MVO.

“Normal portfolio” is a generic portfolio which satisfies Client’s long term objectives & constraints (as formulated in IPS).

SAA aligns the systematic exposure desired. Given the IPS, exposure to different asset classes is determined keeping in view their expected performance.

It did not talk about the efficient market (i.e. asset prices in Equlibrium). CME are built keeping in expected economy & various asset classes performances. If assets price change, u do rebalancing or TAA to caputure advantageous disequlibria within IPS mandate.

ContraCan - What is your point? is it that we are using models that use assumptions that do not necessarily hold in actual practice? Welcome to finance buddy.

You guys are just wasting time on this, but please continue, after all, you are my competition. I’ll be sure to put the “wrong” answer down so long as the CFAI tells me its right. At this point they can tell me to call blue yellow and i’ll do it.

Mark C fail - you got the right answer.

Rahuls - if prices change and assets and asset classes move out of their original allocation, then the only way the strategic allocation can remain the same is if the expected returns etc remain the same. The only way this can happen is in a general equilibrium with random independent price movements. Otherwise one asset’s higher price results in a lower expected return and another’s lower price results in a higher expected return - therefore the strategic allocation should change and not remain a constant.

My point is that there are underlying assumptions to the statements in the course that are not clearly stated, but are assumed to be understood and accepted.


I was trying to say you have you choose asset classes to satisfy your long term return objectives & risk constraints. While choosing asset classes you sync your risk/return requirement + CME for long term.

If assets classes move out of their original allocation then you depends whether you expect them to revert in line with your SAA. If you think they will revert & current changes are only temporiry you may try to capture value thru TAA.

If you think asset classes movement are discounting the changes in CME for the long run you may revisit SAA.

Nevertheless, what i understood from a normal portfolio in macro attribution context to be satisfying long term needs in general. So policy allocation which is very critical & one of the input in discerning Macro attribution factors, is specific to fund sponsor not to the market as a whole.

Your assumption that whether normal portfolio means mkt in equilibrium was the discussion subject here.