Risk Management -- trimmed DFR

This topic deserves a careful analysis and a thorough discussion but its going to have to be over time because there is way too much to cover on this topic for one night.

I use several methods of risk management but I should have more -- over time I will keep refining the process. First method is diversification -- trying to keep positions to a reasonable size so that if any one of them drops sharply, it doesn't ruin the whole portfolio. Today I decided to exercise this part of risk control by trimming my position in DFR by 20%. It was just too big a position size given the risks involved -- I don't believe the mortgage market has suddenly healed itself nor is credit any more available then it was a couple of weeks ago. Investors seem to be worried about how much future dividends will be and whether the merger will still take place or be renegotiated and not very focused on liquidity concerns despite what happened to TMA, H&R block, CFC etc. I have had too big a position and the rebound off the bottom gave me a chance to sell some but I was probably too early -- if the UBS report came out post close, we could see another increase tomorrow based on that analysts bullish remarks (I plan on trying to get a copy of the report tomorrow).

I also try to diversify by type of stock as well as secular theme. My current list of themes:

1. Asia -- their economies are strong and yet their share of the world's value is low -- this will change over time. (matthews asian funds)

2. Asset management -- great way to piggy back on the market's gains with fund flows an outperformance kicker. (AB as well as indirectly with FDS, CME and potentially DFR).

3. DNA/biotech/genomics/etc. -- understanding DNA will have a huge impact on the practice of medicine in the future and there are many ways of playing that (ILMN, LH, GNVC, TECH and sort of MDT)

4. Internet -- we are all spending more time using it and gaining more value from it -- its a huge change in the economy. (GOOG, indirectly MSFT)

5. Energy -- they aren't making more of it; thermodynamic laws make alternatives very unlikely; finding and development costs are rising and that provides floor to pricing not to mention strong worldwide GDP growth. (VGENX, ERF, EPD, MMP)

6. Various smaller themes:
a. miniaturization of technology; growth in digital consumer products -- TSRA.
b. Smart cards -- use of smart card to replace cash in 3rd world economies as the low cost solution (UEPS).
c. GPS -- location based services will be much more valuable over time as applications are built and wireless access improves and the technology becomes more ubiquitous. (NVT).

I am actively looking for additional themes -- my interest in MSM, IHS, etc. is because they look like great companies, but on the other hand, I'm not sure they are beneficiaries of a secular theme I can point to other than energy in the case of IHS.

By type of stock I mean how it is generally characterized -- growth, momentum, value, high beta, low beta, high dividend yield, etc. Not every stock in the portfolio is a rocket like NVT and ILMN. Some are lower beta or pay dividends. Point is that this way if the market is spooked by high PE stocks, my portfolio will experience pain but there will be parts that will benefit too.

Besides diversification I use selection as a risk management tool -- trying to find stocks that are likely to outperform expectations (less likely to decline) that sell for reasonable valuations relative to what I believe they will earn in the future. Yes I am worried about NVT and ILMN's valuations but they are a small part of the total and they have strong fundamental momentum. I think UEPS is very undervalued. Stocks like GGG, LH, MSFT, MDT and AB seem very reasonably priced to me.

I have not taken advantage of volatility as well as I should have -- when stocks go up, I should be more interested in trimming them and also more willing to buy more when they dip. Using selection as a risk control can have merits but its not the best method -- what if I'm wrong about the future? every investor is wrong at least some of the time -- the future is uncertain so there is no way anyone can be right all the time. We can only try to improve the odds of being right. Using selection, you must be quick to sell when the risks outweigh the rewards -- when the upside is limited and the downside is great. This requires diligence -- you have to always be reviewing positions and asking where this stock is on the risk reward spectrum.

By trying to stick with what you know, what you have an edge in, that is how you can improve the odds and practice good risk management. The problem most people -- myself included -- is that we actually know a lot less than we think.

I don't focus on what academics call risk management -- volatility. its not true risk but rather a mathematical approximation. I think the recent past has demonstrated the impact of putting too much counsel in mathematics for risk control.