Sunday Morning Coffee

In the comments of yesterday's post a little discussion broke out about my definition of large accounts and mid sized accounts or more correctly the differences, as I see them, in portfolio construction based on account size.

The relevance could be that many do it yourselfers might be inclined to have fewer holdings not as a function of account size (for our "mid sized" accounts we tend to only have about 20 holdings, most of which are ETFs so that commission expense is not excessive) but as a function of time available to spend on the task. If someone has a full time job then they may not be able to make another full time job out of managing their money which is very reasonable.

I do this going sector by sector same as with large accounts. Decisions need to be made for each sector about overweighting and underweighting and the best proxies (per the person building the portfolio) need to be chosen to fill out the portfolio.

One way to come at this, as we've discussed before, would be one stock and one ETF for each sector which works out to 20 holdings (for someone benchmarking to the SPX). This would not work for everyone of course but it is one way to do it. The idea with one stock one ETF is avoiding over exposure to single stock risk, building in some yield and some precise themes.

I do it a little differently. For the smaller sectors I use just one ETF, again the context is mid sized accounts not large accounts. The three smaller sectors are each only about 3% of the index so even an overweight in those three would still be a small exposure for the portfolio and where commission drag is a concern I think one pick can work.

The financial sector is a tough place for me to find an ETF to use for the way they are constructed. I don't want to own the common of the big US banks, the big European banks, the big UK banks, the big Japanese banks or the big Chinese banks. So for financials I tend use individual stocks that we use for large accounts.

For other sectors we use a mix of an ETF or two and/or a stock or two.

One thing I mentioned yesterday was trying to change the volatility profile of the portfolio with the trade executed. This obviously doable in the context of the type of portfolio we're talking about today. There are a few good compare and contrasts to illustrate the point. In the materials sector you could compare the Global X Fertilizer ETF (SOIL) and the Materials Sector SPDR (XLB); XLB being benchmark beta. The Malthusian argument underpinning SOIL is easy to understand, you may or may not agree with it but it is easy to understand.

Not surprisingly SOIL is quite a bit more volatile than XLB. Someone believing in the argument to own SOIL could keep the fund and switch to XLB at a time where they want to reduce volatility of the portfolio but still maintain exposure to the sector.

There could be a similar dynamic in tech with the eTracs Next Generation Internet ETN (EIPO) Tech Sector SPDR (XLK) although given tech's huge weighting in the SPX, putting that much into EIPO seems very aggressive but you get the idea. With energy maybe this could be illustrated with the Jefferies Wildcatter ETF (WCAT) and the Energy Sector SPDR (XLE).

Before the 1+1=11 brigade comes out, the above symbols are just examples to illustrate a point, we don't own any of them.

The same type of work can be done for yield, cap size, foreign/domestic or anything else you might be interested in trying to capture in your portfolio. More likely there would be several variables being managed at once but again you get the idea.

The picture is from yesterday of Roscoe surveying our second storm of the season.