The Benefits and Downsides of Modern Portfolio Theory
I’ve been reading The New Pension Strategy for Canadians which looks like a decent introduction to investing ideas for non-technical people. However early on the book it starts talking about how Modern Portfolio Theory is the absolute best way to manage your investments. As I wrote about earlier there are some potential issues with using MPT too aggressively.
The book exposes another common misconception that comes from MPT. One of the main ideas is that risk and return are the same thing, since you need to get a higher return for taking on risk and those selling securities will offer a lower return if they have lower risks. Mean people take this too literally and think that it’s a universal rule. It is a useful guideline, since in most cases an investment that offers higher returns has to have some risk for it. However as the original MPT paper stated 60 years ago:
“For any level of risk (volatility), consider all of the portfolios that have that volatility. From among them, select the one that has the highest expected return.”
This is admitting that there are many options that have the same risk but different returns, or the same returns but different levels of risk. If you already have the best possible portfolio and you want to change it, you need to make a tradeoff between risk and return. But if you have a bad portfolio (and there are many many of these) then you can increase returns and decrease risk at the same time.
Aside from that the author highlights one major benefit from the spread of MPT: pension funds moved around that time from looking at individual investments and managers to thinking of the portfolio as a whole. This is always a good move. I track our investments as one combined portfolio to see the real allocations and total returns, and the returns of any single investment or account are mostly irrelevant. Compare this to the online discussions a few months ago about returns in 2012 where people would say “I made 70% in my TFSA and lost 20% in my RRSP” – if their RRSP is a much larger account and important to their future finances, that’s a portfolio that did poorly.
Another good point in the book is that although short-term volatility is a poor measure of risk, many pension funds measure long-term volatility which is much more important. Knowing that my portfolio could return 2 – 8% per year over the next 30 years is different from knowing that it could return -6 – 14% per year over that time. There are still limitations to what we can predict but at least applying the volatility measures over the right time frame gets closer to the right answer. I’m not in the pension world so I’m not sure what percentage of funds measure risk in comparison to their true liabilities. Hopefully most do.
Overall Modern Portfolio Theory seems to have introduced some useful ideas. Those good ideas can be used without the precise and sometimes overly-sensitive math. Many investors and mangers are probably better off using it rather than the alternatives though, as long as they understand what it can and can’t do.