Ridley (TSE:RCL) has been a star performer in my portfolio this year. Only 7 months after I purchased it for $12.45 I have sold it for $24.32. So that’s almost a 100% return right? Well not quite, the currency moved against me by 13% which knocks that 100 down to more like 70%. Ah but that’s where my hedge comes in, it should make up the shortfall – that’s what I thought. Well imagine my surprise when it contributed just 5% to my return. I somehow screwed myself out of 25% of return!
So in the end I made a lousy 75%, or 130% annualised. I should be happy, but add this screw up to the fact shares have risen another 10% since I sold and I’m not.
Just a note on the sale before I get into my hedging screw up, it wasn’t actually an intentional sale, I had put a limit order in a couple of months ago at my target price and it just got hit as the company announced a $2 special dividend. I’m sad to see it go, it is a strong business that I originally wrote up on Seeking Alpha. It’s now in the pro library so for those that can’t read it, my valuation was based on a high margin segment which was being obscured by poor performing units. I valued it in the range $226m – $311m and today’s market cap is $320m. Earnings are up 8% in FY 2014 which is good performance. It is also majority owned by Fairfax, a famous value oriented firm which means it is well run and capital is managed well. It’s now trading at a forward P/E ratio of just 15.4 which I consider reasonable so that is why I’m a bit sad to say goodbye to this stock. I have a feeling I will look back at this stock in 10 years and say “why on earth did I sell that”. But conversely I am happy portfolio cash is now at 15%, giving me more firepower.
So let’s take a look at exactly how I hedged. I use spread betting, which means I bet on movements in the GBP/CAD exchange rate and get paid a fixed amount per pip (I use this because in the UK is it cheap yet highly leveraged, so requires little capital for the hedge – but it is illegal in the US). I sized the hedge against a 5% move, so I simply looked at the CAD$ value of my shares, roughly $5,000, worked out a 5% move against me would cost me £130, then sized my bet so that I won £130 if it moved 5% (which was 14p per pip). It isn’t actually that sensitive to the size of the move. For example doing the same sums for a 10% move still gives you a suitable position size of 14p per pip. So where did it go wrong?
Well let’s look at how my trade actually worked in real life.
Opening trade: $4,942 / 1.6322 = £3,028
Closing trade: $9,655 / 1.8573 = £5,198
Change in exchange rate: +13.8%
Under my model, the hedge wins (1.8573 – 1.6322) * 0.14 * 10,000 = £315.
In CAD$ my trade made 9,655 / 4,942 – 1 = 95%. If exchange rates didn’t change I should have made £3,028 * 95% = £2,876 profit – but in reality I only got 5198 – 3028 = £2,170. Hence I’m short £706, not £315. What I should have done was hedge against the final sale amount rather than the starting amount, that way I should receive a full hedge.
But that creates a problem doesn’t it. How do I know what price I will sell a share for? Is it even possible to predict with any accuracy what price you will sell a share for. Of course I know my estimate of intrinsic value and target price, but if I am wrong and end up selling at a loss, then I will have over-hedged and could compound the losses!
Take for example my Emeco Holdings position in AUS$, which I aim to sell for triple its current price. Hedging triple my AUS$ exposure is a big difference and the hedge can suddenly lose me money instead of offsetting gains elsewhere.
There is also the time element. If I plan on selling a share in a year or two at double the price, then it may make sense to hedge against my target amount. But if I have a long term holding like Berkshire Hathaway, that I plan to hold for 10-20 years, its intrinsic value is going up over time and it would be absurd to predict its value 20 years in the future and hedge against that for the next two decades!
There is also another problem. Currency futures are the cheapest way to hedge, even with spread betting. But they only last 3 months which means I adjust position size every 3 months when I renew the bet, based on the most up to date value of my holdings.
Let’s say I have $5,000 of holdings and 3 months later it is $10,000, then after another 3 months is $5,000 again. At the same time the currency hedge gains 5% in the first 3 months, then loses that again in the next 3 months to go back to where it started.
My start and end values are the same, and worth the same in $ and £, yet because my second hedge was on $10,000 after the temporary rise, my hedging returns are not 0 and will actually be negative.
This is less of a problem for short term holdings, but can make a big difference on long term holdings which will vary in value undoubtedly.
I don’t have an easy solution to this. I expect most of my hedges to “make” me money over time (or I wouldn’t bother hedging) so it is important for me to get the position sizing right.
By buying a stock in the first place I am expressing a commitment that the stock is worth at least a certain amount, so part of me thinks that the rational thing to do would be to hedge as if I am correct and just accept the added risk it brings. If I’m in doubt surely I shouldn’t have bought the stock in the first place!?
Yet when positions like Emeco require me to triple hedge size and increase my volatility I can’t help but hesitate. I’d be interested to know how others size their hedges.
Disclosure: Author is long Emeco, ASX:EMCO and has no position in Ridley Inc, TSE:RCL