The Shape of Things to Come pt I

Churchill was responsible for some cracking quotes. ‘We shape our houses and then they shape us’ is, er, quite incredibly good. I read another while buying my mum a present online. ‘Gentlemen buy their cheese from Paxton & Whitfield’. It shows that even the greats can have an off day.

Charlie Munger is one for more plain spoken remarks. His view is that the key to investment is not having a high IQ, it is making your IQ count double. He breaks down four key attributes; constant learning, running multiple mental models, patience or discipline, and the willingness to make big decisive investments when the time is right.

I’ve spent quite some time over the past few years thinking on how I could best construct a fund – to shape my trading to fit these qualities.

I knew from the start that it would never be a long only relative performance fund. I know some real pros that do this. But for me, if I think that the market is going down, I want to be short, not just defensive.

And I knew it had to be global and it had to be multi-asset. There is little point sitting in Japan in 1992 if you think the opportunities are in the States. Or buying gold stocks, when you would rather buy gold.

But the big issue is how I structure the fund; the house that shapes my performance. And that comes down to where I’ve got an edge. There are four or five things I reckon I can do that help me make money;

- Wait patiently for a high returning business to generate strong shareholder returns. I’ve held tobacco stocks and GEA for four years through thick and thin.
- Identify big themes early. I was long gold and resources from 2003. I was short Irish banks from November 2006 (yes, sometimes I’m too early).
- Cut when I’m losing. I run specific cash losses for any position; a loss of one percent of capital for long term stocks. A quarter to one percent for leveraged positions.
- Identify self reinforcing macro themes across several asset classes.
- Spot when a market is turning from a good to a bad risk/reward bet.

And that’s about it. So what I’ve done to try to ‘outperform my IQ’ (yes, perhaps not a difficult task), is to run two funds in parallel.

The first is the base equity of the fund – and that’s what I’ll talk about today.

The idea here is to use the 100% of base equity in the fund to distribute between cash, equities, bonds, credit and commodities. And by optimum, I don’t mean putting it on some bizarre, volatility adjusted, efficient frontier straight out of modern portfolio theory. I mean the blend of assets that will make me the most money over the next several months. The assets with the best risk/reward. The plan is to move the asset allocation around no more than, say, four times a year. Right now I’m 85% equities (more on that later), 5% cash, 5% Gilts and 5% gold. Last time I moved was mid-late March, when I sold a 20% position in Gilts for equities.

The next issue is what I hold within each asset class, and particularly equities. Now it’s common knowledge that, on average, equities are the best asset class to hold – on account of their risk premium. If you’re prepared to take the risk – the volatility and the potential loss. Equities tend to pay 4% or so more than bonds, and say 5% more than cash. That four or five percent is the risk premium. And when you compound those high returns, on average, you double your nominal money every seven years, compared to every 14 years for cash. And you go from 2x to maybe 3x faster than cash if you’re thinking in real terms which, of course, you should. Buffet’s quote; ‘I’d rather make a volatile 15% than a steady 12%’. Compounding is why.

And you can go a lot faster than the 8-10% average for stocks in two ways. First, if you walk away from stocks when the risk premium is low – say below one or two percent – and go heavy when the premium is high – say six or seven percent or above – then you compound at a much higher rate. That’s harder than it seems – you might have to wait out three or four years (you did in 1999). But it’s worth it. Compounding at 12% rather than 8% doubles you in six years rather than nine.

And then there are the themes. Oftentimes individual sectors and stocks will offer a poor risk/reward. Others like infrastructure, energy and mining now price in declines in prices and margins, and a slowdown in growth. If my theme suggests the opposite, which it does, then the notional risk premium – what you’re being paid to take the bet – is much higher than most people believe. You may find yourself compounding at 20-30%. This isn’t what Buffet and Munger do – the look for the moat – the long term sustainable competitive advantage – of an individual business. I’m looking more for thematic medium term calls – say three or four years. If I’m right and resources and infrastructure in general have got a large moat over the next few years – then the high returns that these companies are making will compound over that time.

So I manage my equity calls based on what I think are three year themes. I generally want to hold a stock that long – provided I don’t lose 20% from entry, or the risk reward gets too far out of whack.

And I make sure that the equity bets are big enough. Each position starts at 5% of fund value (NAV), and I double up if the fundamentals and the chart look right. Right now I have three stocks in which I invested a full 10%; Patterson (The US natural gas driller), SSAB (The steelmaker), and GEA (The German Mittelstand conglomerate). And then I have a bunch of others that started at 5% including ENRC (Kazakhstan Ferrochrome), ABB (Swedish power infrastructure), Ocean Rig (Norwegian deep sea drilling), Cosan (Brazilian sugar and ethanol), CVRD (Brazilian iron ore) and Petrobras (Brazilian oil). And among my smaller positions, I bought some Crox two weeks ago at $11. I know, that’s a bit leftfield, I’ll explain it later.

As you can tell – I’m not big on diversification. I’d rather hold cash than a stock I don’t believe in. Nor am I big on VAR. There are three reasons. First, it’s based on the assumption that the world is ‘Gaussian’ – that all risk is distributed on a normal curve. But Benoit Mandelbrot – in his book ‘Market (Mis) Behaviour’ – showed that risk is distributed on a power law. So you get fat tails – a much greater likelihood of extreme events.

Second, it usually takes volatility from the recent past (Most hedge funds use six months) as a guide to future volatility. But that’s misleading. Hyman Minsky showed that ‘stability breeds instability’. Something that the recent credit crunch demonstrated rather clearly.

And third, VAR induces you to do exactly the wrong thing. When markets have fallen, investors become more fearful. Implied volatility (The VIX) rises. Which means that the risk premium – the prospective returns over cash or treasuries – must be higher. Your VAR goes up so your friendly risk manager tells you to cut your positions. But as I talked about earlier, in the discussion about compounding returns, it’s precisely when risk premia are high that prospective returns are high, and that you should be investing most. It also works in reverse. When the VIX and VAR is low, risk managers induce you to bet more, just when you should be taking risk off the table. VAR is for the birds.

So the aim is to hold stocks long enough for an oversized risk premium to compound. It’s where the patience and discipline comes in. I try to check the stocks only once a day. I review them only if they’ve fallen 10%, or risen 25%. And I think that gives me an edge. Tomorrow – the shape of things to come, Pt II – the leveraged portfolio….