Hello everyone, welcome to the second week of my class and international portfolio management. This week we're going to start with introduction into breeds in Hood Deboer portfolio, return frame analysis framework. This framework allows you to understand major sources of portfolio performance, what actually could be those sources? Well, first of all, we understand that we can select some instruments in our portfolio and. Return of this instruments would be result of our portfolio. This is the simplest case of sources of return, which is called returns from security selection. Now we also know from our previous video previous week that there is such thing as investment policy statement, investment policy statement shows you major allocation between major asset classes allow investable or allow it to invest by your message. And this is returns from strategic asset allocation, and we also known for our previous video that investment policy statement could also allow you to deviate slightly from strategic asset allocation if you expect some particularly good or some some particularly bad conditions for different acid classes in future, and you believe that this information is not already in prices, so these deviations are called tactical asset allocation. So the interest between interest to strategic and tactical style location had started in ATS of 22nd century, probably from the paper or probably due to that fact. Prison, who borrowed, published their paper which is called determinants of Portfolio Performance. 20 years later, one of them would right? We would not have gas in that six page article will be the focal point over 20 years discussion. Basically, their conclusion was that investment policy statement explains up to 936. or for two variance of UK pension plans. So what was the methodology using which they came to such a conclusion? They have calculated 4 magic numbers. The simplest one is quadrant 4# or. Actual portfolio return this is weight of your actual weights multiplied by your actual return. It's in every asset class. Now the second number is watch a tool you would have is if you would just stick to your policy and give up the security selection and market timing or policy return is just policy weights multiplied by benchmark returns. So these are two vectors multiplied by each other. Now we have two intermediary numbers, quadrant 2# is policy and time return or actual weights multiplied by benchmark returns and we also have policy and security selection which is policy weights multiplied by actual returns. Now we are making the following simple manipulations with these four numbers. So first of all we take. Policy in time return Anns obstruct policy return from it and we obtain timing return or return due to timing. Then the same thing with policy and security selection and policy return, so policy and security selection minus policy return is selection return and finally we take actual portfolio return we took. We take policies, abstract policy and security selection policy and time will return from it and with summit with policy return. To capture all other effects an as I've said, the majority is currency, affect. Totale actual portfolio. Two or minus policy return is your active return, so this is decomposition of your active return. Now you probably have heard about the story of Harvard Endowment. This is one of the largest endowments in the world with one of the best quality of management. And yet they suffered huge losses, almost two billions in investment returns. Over year 2016, let's try to apply Brinson who Bieber methodology to understand what was the sources of such underperformance. So you remember that we have to prepare for inputs, vector of policy, portfolio weights, vector of actual portfolio weights, vector benchmark returns an actual returns and the number of elements in those vectors will be equal to the number of asset classes which are allowed by our investment policy statement. When we are working with the Harvard Endowment reports, we understand that. The way higher Harvard is reporting about their strategy is not exactly compatible with the framework of Britain who Bauer, so we have to make certain assumptions. Their policy portfolio is changing overtime, so we I take the long term average policy portfolio as a policy portfolio. We also do not have policy information for the year 2016, 2016, so. We have on the minimum maximum allowable weights, so I took median between these two numbers then. We are normalizing all that numbers to sum to 100% an we obtain policy portfolio. Now if we would substract these longterm- average portfolio from our active allocation for the year 2016, we would have obtained their active weights. You see that they were underweight in natural resources, absolute returns and foreign equity. Probably expecting something bad in those sectors they were overweight. In realestate, private equity and emerging equity. Probably expecting something good in those sectors. Finally, let's look at their returns. So the difference in returns between benchmark an actual return would be due to instrument selection, so they've managed to be the market only in two sectors versus real estate, and the 2nd is private equity in all other sectors they was. There were other worse than the market or equal to the market. Now if we would apply the brakes and who would be borrowed methodology to their results, we would obtain the following. We have those four numbers according to their methodology 100, Four, 321 and then tactical asset allocation or quadrant 2# minus quadrant number one is point 5 so they were actually we actually had positive return. In tactical asset allocation, they have also positive return in other categories, so the affect of currency reprising in other factors was positive for their portfolio, but they had extremely negative return. Insecurity selection minus 25. which resulted in total active return of minus 11. and that actually contributed to this extremely bad year for Harvard Endowment. So this is how Howard, how Brinson, who borrow methodology, could be attributed for demonstration purposes only. Of course, to results of Harvard's endowment. If you need more simpler description of Prince in Huber approach more in detail description, you could refer to the a longer video in our supplements.