Introduction to Asset Allocation & Portfolio Construction

Asset allocation and portfolio construction is incredibly easy - and fun, too! In this section, we will feature the basics of asset allocation and portfolio construction, and define a few easy steps for you to understand and implement it by yourself.

Introduction

  1. What is asset allocation and portfolio construction?
  2. Why do professional money managers use asset allocation?

Basic Concepts - What information do we need to construct an optimal portfolio?

  1. How do asset classes relate to each other?
    1. Basic Terms: Covariance and Correlation
  2. How much do I earn with each asset class and what risks do they pose?
    1. Basic Terms: Historical Returns and Volatility

Introduction

What is asset allocation and portfolio construction?

Asset allocation and portfolio construction are two terms used to describe where you allocate money (asset allocation) and how you construct your portfolio (portfolio construction). As such, the two terms are very much interconnected.

You can differentiate between the time frame you invest, such as that strategic asset allocation usually refers to your general allocation over a multi-year time period (e.g. you decide to generally allocate 30-50% of your wealth to equities, while you allocate 50%-70% to bonds). Tactical asset allocation refers to the allocation on shorter time periods, such as when you decide to move a higher proportion of your portfolio to equities (for any given number of reasons) for the next couple of weeks or months.

Why do professional money managers use asset allocation?

The reason why money managers use asset allocation is simply due to the fact that financial market returns are unforeseeable. As a result, and according to classic financial market theory, a well-diversified and well-structured portfolio mitigates your investment risk.

Suppose you have a classic portfolio of 40% stocks and 60% bonds. Let's further assume you start your portfolio with 100 USD in the year 1980. In that scenario, you have an allocation of 40 USD to stocks (equities), and 60 USD to bonds (such as US government bonds). Both stocks and bonds have different trading patterns, and they usually do not trade 1:1 in the same magnitude or direction together. The relationship between them (or any asset class with another one) is referred to as covariance, which describes what happens to one asset class (e.g. equities) if the other asset class (e.g. bonds) is up or down on a given trading day. If, for instance, stocks go down tomorrow by 1%, the other side of your portfolio (bonds) usually do not go down by 1%, too. Most frequently, bonds and stocks move in opposite direction, meaning that bonds go up when equities go down and vice versa. So in our case, you have lost 40 cents on your equity investments (40 USD x 1%), but the bonds might have gone up by 0.5%. In that case, your total portfolio change is -10 cents (-4 USD + 60USD*0.5%), much less than if you had just invested straight out in equities. As such, asset allocation mitigates risk and leads to straighter and more foreseeable returns, which is especially necessary if you manage a large amount of money and want to preserve and grow it over time.

Basic Concepts - What information do we need to construct an optimal portfolio?

How do asset classes relate to each other?

A key point of asset allocation is to understand how the different asset classes (again, this could be stocks, bonds, real estate, gold or any other security type) you hold in your portfolio relate to each other. Two terms are especially important in this context: Covariance and correlation.

Covariance

Covariance shows us how asset class 1 (equities) relates to asset class 2 (bonds). The question we answer is: If equities go up, do bonds go down or up, too? However, do to the mathematical nature of the covariance formula, we do not see the strength of this relationship, so we do not know by how much bonds go down or up if equities go up. For answering this question, we would have to look at the correlation of the two asset classes. Before we go into that, you can have a look at the covariance of selected asset classes here:

The information we can take away from this is, as an example, 7-10 year US government bonds generally move in the opposite direction of equities (S&P 500) or that corporate bonds generally move in the same direction as equities (S&P 500).

Correlation

As said above, correlation will help us answer the question of how many percents asset class 2 (bonds) go up if asset class 1 (equities) goes up (or down, for that matter). The correlation coefficient is a numeric value between -1 and 1. A value of -1 means that that specific asset class goes down the same amount the other asset class goes up (they move exactly opposite to each other). A value of 1 means they move in exactly the same direction and magnitude. You can have a look at the correlation overview here:

Here, we can take much more information away with us. If you look at the row S&P 500 (which are equities), we see that they generally move in the same direction and in comparable magnitude as real estate. Similarly, we can see that there is a negative relationship between equities (S&P 500) and government bonds, since they generally move in the other direction. We can also see that the S&P 500 and corporate bonds are generally uncorrelated (meaning there is no special relationship between them as the correlation coefficient is close to 0). Generally, the more the correlation coefficient between two asset classes goes towards -1 and 1, the stronger the positive or negative relationship is.

How much do I earn with each asset class and what risks do they pose?

Historical Returns and Volatility

The other two very important variables we have to look at in order to create an optimal portfolio for us are historical returns and volatility. Historical returns tell us how much an asset class has returned to its investors over a previous period (this can be anything from a day to several decades or even centuries). Volatility, also called standard deviation, tells us how much an asset class moves. Generally, the higher the standard deviation, the riskier an asset class is considered to be (imagine you invest 100 USD in a particular asset class and tomorrow it is worth 120 USD, while the day after is it worth 80 USD and then again 115 USD - that is called high volatility).

Below, you can find the historical returns and volatility, expressed as percents since 2002:

Now, you can see that the different asset classes here both have returned not only different profits to investors over time, but also different volatility. The best returns were made in real estate with an annual gain of 10.16% and a volatility of 31.94%. Second to that are equities (S&P 500) with 8.62% and a standard deviation of 19.18% per year. The Sharpe ratio is the trade-off between return and risk (volatility or standard deviation). The higher the Sharpe ratio is, the better it is and the more you earn for any given level of risk.

8.62% certainly sounds good, especially if you receive it over an extended period of time as you can double and even triple your money if you let it be invested long enough. Nevertheless, the standard deviation is a key concept to put focus on, as you can see below when we look at the chart of the S&P 500 since 2007:

You can see, the returns have been spectacular (not even considering reinvested dividends), but there are period when the value of your investment falls sharply, such as in 2007. If you have saved up money all your life, and want to avoid such periods and see your savings diminished, asset allocation is what you want, because we can minimize and/or avoid these drawdowns and you'll sleep much better - and that even without having to accept lower returns.

Earning as much as you can for a given level of risk, is the key pillar of portfolio construction.

We now have everything that we need to construct a portfolio: How asset classes relate to each other (covariance, correlation), how much they return and how much risk each one has.

In the next section, we will see how we can combine these different asset classes to construct a portfolio that gives us a good return at a much lower standard deviation than any of the asset classes alone.