Part I

Ever hear music that uses computer synthesized sounds? Of course you have, if you’ve heard any made since about the middle of the last century. It would be hard to find music made since the turn of the century that didn’t. Almost all recorded music is now produced in computer based studios, and they make extensive use of synthesized sounds, both those that sound synthesized and those that sound like mechanical instruments.

There are several computer based synthesis methods. Two of them are complementary approaches, additive synthesis and subtractive synthesis. They are related according to a mathematical theory called Fourier analysis. Now, before I lose the majority of readers, the essence of it is that any periodic waveform – any musical tone – can be constructed as a sum of sine waves of differing frequency or period and amplitude. One can synthesize any such waveform by adding them together. One can also start with all of them and subtract those that don’t belong to the waveform being sought. The Fourier transform is that piece of mathematical magic that ties them together.

Though these approaches are conceptually or procedurally the opposite of each other, they can turn out equivalent results … you can construct the same waveform using either method. Although you can get to the same place by either route, depending on what waveform you want to construct, usually one approach is simpler than the other.

It turns out that synthesizing an investment portfolio is much the same proposition.

In additive synthesis, you specifically select from the universe of available investment assets those you specifically want, like a painter applying paint to a blank canvas. The complementary approach is subtractive synthesis; you start with everything and subtract what you don’t want, like a sculptor starting with a block of marble and chipping away everything that isn’t the sculpture.

As with making a musical waveform, you can arrive at the same result either way. But which is the more direct route depends on where you want to end up.

In investing, this translates into your set of priorities and whether you have a clearer notion of what you want or what you don’t want. They are also not mutually exclusive. You can allocate part of your portfolio to additive synthesis and part of it to subtractive synthesis.

The additive approach is the more intuitive and straightforward. What could be simpler than buying what you want? It requires no justification.

The subtractive approach is a bit more nuanced. It starts with the notion that you don’t – can’t – know what your selected investments will return, but that it is possible to know what risks they present.

With this approach, you start with the Global Market Portfolio, and remove or underweight what you don’t want or want less of. The GMP is the default portfolio … the totality of the world’s assets – what you would own if you made no selection at all, and so represents zero risk due to your selection of investments. Because of this, any deviation from it directly represents a defined risk, so you know exactly what your risks are. It also forces you to consider a priori the entire menu of available investments, not merely widely promoted ones.

If for example you know you don’t want low or no yielding teracap stocks, or low quality or richly valued stocks of any market cap, you can buy them all except for those stocks. You then know that the risk you are taking is that the these types of stocks will underperform over your investment time frame. That, for example, is a risk I’m willing to take, so I’ve devoted much of my portfolio to it, for example via the Income Portfolios listed on the Financology Model Income Portfolio page.

Another thing I don’t want is bonds with credit risk or currency risk, because “fixed income” isn’t really fixed for me if it’s not fixed in dollar terms. I want my bond positions to provide some overall nominal stability by acting as a counterweight to the nominal risks in my stock and commodity positions. This way I know exactly how much nominal risk I’m taking. So the only bonds I own are US Treasuries.

Another part of my investments are specific assets that I want to invest in just because I like them. I haven’t discussed most of them in these posts, especially the smaller cap assets, because they’re specifically tailored to my interests and priorities, not necessarily of general interest to readers. This is the additive synthesis part of my portfolio. A more accessible example is in a recent post by JK, updated in a comment on my last post. You’ll want to read it in its entirety, but for purposes of this discussion it’s an excellent example of a portfolio constructed by additive synthesis.

The point of all this isn’t to advocate any particular approach, but rather a systematic way of thinking about the portfolio construction process. Are you practicing additive synthesis? Subtractive synthesis? Or some of each? At least knowing which or how much of each you’re doing allows you to make it not an accident but a deliberate decision and helps clarify how to translate what you seek into a real world portfolio.