SIF Portfolio: Diversification Dilemma Makes Me Question H&T

30th Aug '17 by Roland Head

8 comments 6596 reads

I’ve been struggling with diversification this week.

The SIF portfolio is now reasonably diversified, but is running short of cash. There’s not really any compelling reason to add more stocks without selling some first. But there are a few companies in my screen results I could add if I wanted to.

The problem is that a rules-based portfolio isn’t about doing what you want. The selection of stocks is meant to follow a process. So what do my rules say about this situation?

When I originally launched the SIF portfolio in April 2016, I laid out the following guidelines for diversification:

The target size for the portfolio will be 15-20 stocks.

Macro-economic diversification will be managed by maintaining an approximate 50/50 split between defensive and cyclical stocks. Beneath this, I will try to ensure that at least four of the ten main stock sectors are represented.

Here’s how I’ve managed against these three criteria:

Portfolio size: Reinvesting profits and extending my holding period from six to nine months means that the portfolio currently has 22 holdings. There’s only enough virtual cash to add one more stock.

Cyclical versus defensives: I’ve discussed the portfolio’s shortage of defensive stocks many times. The market continues to price most defensives too richly for my screening criteria. At the moment, only about 15% of the SIF is in defensive stocks.

Sector diversification: The portfolio contains companies from seven of the ten main stock sectors, so this requirement is satisfied.

The challenge facing me now is to decide how to proceed from here.

I may need a new rule

My first choice would be to add another defensive stock. But there aren’t any new defensive stocks in the screen results at the moment, so this isn’t possible.

Another possibility would be to add a stock from one of the missing sectors (Energy, Healthcare and Telecoms). But none of these qualify for my screen at the moment, so this isn’t possible either.

This leaves three possible options:

Add new stocks to sectors where the portfolio is underweight, in order to try and achieve even sector weightings.
Ignore sector weighting and focus on adding the companies with the highest StockRanks.
Don’t buy any new stocks at all. The portfolio is already over its target size. I can save the limited cash for the next defensive opportunity.

My original rules don’t really…

Unlock with Email

Unlock with Google

Already have an account?
Login here

Nick Ray

30th Aug '17

Do you have rules for when to remove stocks from the portfolio? Superficially, it looks as if you might be approaching a situation that happened to me, where I had about 25-ish stocks and it was getting difficult to decide what else to add. But then I realised that perhaps only about ten of the stocks were really performing anyway and a better approach would be to trim away the dead wood and bring the portfolio back to maybe 15 stocks that had earned their place, and then start growing again from there.

Roland Head

Hi Nick,

I didn't mention selling in this piece, but I do have a rule for that too.

Stocks are held for a period of nine months and then reviewed. If they no longer qualify for the SIF screen, they're sold.

If they still qualify for the SIF screen after nine months, they remain in the portfolio on a rolling monthly review until they fall out of the screen, when they're sold.

Hope this makes things a bit more clear.

Regards,

Roland

Howard Adams

Roland

I have two points to suggest.

The first echoes Nick Ray's suggestion about trimming dead wood.

Obviously this would create new funds to allocate to better performing options. But most importantly, as I have found with my own portfolio, it can really remove some headwinds against achieving a good portfolio performance. In the first instance, after a gain of say 10%, which then suffers a fall of 10%, it requires a gain of 11.11% to get back to parity. That's an extra headwind. But, in the second instance if you have reallocated those released funds to a holding which is performing more strongly than the original, you will recoup the lost 10% more quickly.

My second suggestion is that you look at your portfolio from 27th March 2017 to see what your defensive/cyclical balance has been since then and then use this as a guide to select which option you are considering. Because, its from this point that your SIF really takes off against the FTSE All Share. It seems your mix really worked well for you.

Regards
Howard

The other thing you could do is to rebalance by top-slicing your best performing stocks. This will free up some cash for new candidates, as well as taking a profit without exiting from good performers.

Top-slicing is a form of volatility-harvesting which I find works really well.

Hi Howard,

Thanks for your feedback. I have changed my selling policy this year, in that I've extended the minimum holding period from six to nine months. I don't currently have any plans for further changes, but this is something I think about regularly.

Your idea of selling losers if they fall more than a certain amount could provide an additional line of defence against big losses. Avoiding these is one of my top goals -- after all, a 50% loss requires a 100% gain to recoup.

As regards the composition of the portfolio since 27 March, the answer to your question is that the defensive/cyclical mix has varied between one and three stocks during that period. But none of these have performed very well yet -- McBride (LON:MCB), Carr's (LON:CARR), Fulcrum Utility Services (LON:FCRM) are showing a small aggregate loss at the moment.

Despite this, the portfolio as a whole has performed well since March. So the cyclical stocks must have delivered all of these gains and cancelled out the underperformance of the defensive stocks. This seems to confirm that in a bull market you can pay a price for caution. I'm hoping that the payoff is greater downside protection when the next market correction comes along.

Regards,

Roland

Hi Nick,

I agree that top slicing can be a useful technique, although I think it's best used relatively sparingly in order to minimise transaction costs and to avoid the temptation to do too much tweaking.

For the SIF portfolio, my rule is that I will top slice a stock of the price doubles. Otherwise I'll leave the holding untouched until it's due for its nine-month review.

Regards,

Roland

Roland

Thanks for the response.

Interestingly I have found that I have become increasingly preoccupied with 'selling losers'.

Clearly the Stocko tools help us all to find stocks which perform. Ed's NAPS/SNAPS is all about increasing the probabilities that performers will outweigh non-performers.

As this probability biasing approach began to dawn on me I challenged myself as to how I could watch and react to falling stocks fast. I now work twice as hard at watching the bottom end of my portfolio than I do the top end.

My positive performers look after themselves, or get added to when moments are right.

My losers get dropped very rapidly, either as complete exit or as 50% sliced. If they can counteract these examinations then they are held, with a heightened watch applied. On rare occasions they can become recipients of new funds.

To me this is simply tipping the investing efforts in the right direction. Stocko makes it effortless to identify buying options. Buying holdings is effortless. If they are good holdings they will go up effortlessly. But, if they are not they will not, thus as an investor why not redouble efforts, application of tools and application of brain power into stopping downsides.

It's only my view though, I'm sure others will not agree.

Regards
Howard

Hi Howard,

I agree that losing stocks need close attention. One point I'd make in addition to those you mention is the opportunity cost of holding under-performing stocks for extended periods. This becomes apparent if you calculate the annualised total return on your investments, rather than simply the total return.

For example, a 50% gain over four years is equivalent to an annualised return of 12.5%. But over a period of eighteen months, it's equivalent to an annualised return of 31.5%. That's a big difference.

Although it goes against the grain for value investors, it can be more profitable to ditch an underperforming stock after (say) six months and then buy it back again a year later, rather than sitting on the shares the whole time.

The other way to improve annualised returns is to average down into losing stocks. This requires a higher level of conviction and carries the risk of throwing good money after bad. But it does mean you don't miss the start of the recovery.

As you say, everyone has their own approach to managing losses. I think what's essential is that as investors we do something. The 'rabbit in the headlights' approach is the least likely to work.

Regards,

Roland

SIF Portfolio: Diversification dilemma makes me question H&T

I may need a new rule

Unlock the rest of this article with a 14 day trial

About the Author

Roland Head

8 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...