The question of how large a position to take in various securities has generated a lot of literature in financial academia. I won’t bother repeating it here, but specifically, the analysis deals with how to mitigate systemic and non-systemic risk, and diversification is one method to reducing non-systemic risk.
The concept is pretty simple – if all your eggs are in one basket, if you drop that basket then you are hooped. But if you have two baskets then you at least still have half your eggs left in case if you trip.
I noted on Gannon on Investing (somebody that writes very well and comes from a similar background as mine in Finance, mainly through self-teaching) that he believes in big concentration. Specifically, he wrote the following:
My Most Controversial Investing Belief: Extreme Concentration Works
I buy very few stocks. Again, this comes from personal experience. By far, the worst losses to my portfolio came in years where I held the most stocks. The best performance came from 25% or bigger positions in my portfolio that I chose to hold longer term.
I’ve made a lot of money by:
Sticking around for the buyout
And having more than 25% of my portfolio in the stock when that buyout cameToday, I would never buy a stock that makes up less than 10% of my portfolio. I prefer not to start a new position unless it is expected to eventually be 25% of my portfolio. I am not interested in owning more than 5 stocks at a time. I’ve done it – like with Japanese net-nets. I may do it again in similar basket type situations. But I’m a lot less likely to. So, basically I own 4-5 Warren Buffett type stocks (in terms of competitive position) bought at Ben Graham type P/E ratios.
Right now, I’m looking for a European stock to put 25% of my portfolio into. It won’t be 5 European stocks at 5% each. It’ll be one at 25%.
I will distill this principle in another manner: an individual investor can only spend so much time researching securities. Very frequently this process can be as laborious as going through a hundred companies and then throwing all of them on the watchlist and not do anything because they are trading at incorrect valuations. When you do finally have a “hit” that reaches your hurdle rate with an appropriate margin of error, why should you take something that you’ve deemed to have a good risk-reward ratio and dilute your own investment research with a small position size? Good opportunities don’t come often, so why not pounce on it with two feet when they occur?
Personally, my own position sizes are determined by my own perception of risk and a slight modification of the Kelly Criterion. If the risk is high (which would require the reward component to be extraordinarily high), then you keep your bet size small. If the risk is low, then you can afford to concentrate without incurring a blow-up. Probably the best example in my own investment history was putting a large position on Rogers Sugar (TSX: RSI) back during the middle of the economic crisis. It was a classic example of a low risk, medium reward type situation where you can afford to put capital to work.
Ultimately estimating the risk is quite an art because you also have to factor in risk that you never would have conceived of. For all publicly traded companies, such risks include management fraud, accounting misstatements, an earthquake ripping through the corporate headquarters, a plane crash with the executives on board, etc. This alone is probably a good enough reason why I wouldn’t go far above a 20% position size.
That said, if that starting 20% position doubled in your portfolio, it would become 33% of your portfolio. Learning how to pare concentrated positions that have grown due to appreciation is a skill. While I have tended to have decent skills on my entries, I still haven’t mastered the art of selling profitable positions – I typically am one to leave the party early.
Finally, the academic rules concerning position sizes tends to revolve around the management of large portfolios rather than individual investors applying their niches on the markets. The rules of diversification change when running a billion dollar hedge fund versus a relatively smaller portfolio.
Given the high amount of cash in my portfolio, it is always tempting to allocate more cash than usual into new positions, but that is a sure-fire way of losing money.
I agree Sacha. Position sizing and paring are two critical actions which get scant attention in financial literature. My own solution to these problems is to tie my maximum position limits to the passive income generated by my portfolio. When positions exceed these limits, I pare them back (if my thesis is intact) or sell them out.
This approach gives my portfolio a self- healing quality. For each investment, I assess the risk and ask myself how many months of passive income I am willing to forgoe to repair the potential loss. For example, a junior mining stock represents a 100% potential loss. The likelihood of the loss is not remote so I would want to recover the loss in 3 months. Therefore, my position limit is 1/4 of my passive income or 1.8% of my portfolio at the present time.
This approach also forces me to be disciplined about maintaining a strong passive income component to my portfolio. Put another way, it keeps me focused on growing the productivity my portfolio rather than simply its size.
You make a good point about one’s income and position sizing. For somebody starting out with investing, I actually have nothing wrong with them investing 100% of their portfolio into something (i.e. hopefully a well-informed decision), as long as it is a modest fraction of their annual income.
As the portfolio size grows larger than one’s annual income, then much more prudence is required with respect to sizing.
There is some psychology involved. Whenever investing in something and it goes correctly, one always gets the feeling they should have put in more. But I try to buffer that against the thought of “What if this sucker blew up on me?” and then look at the relative loss in the whole portfolio.