Selling BHP Billiton after its recent share price gains

I added BHP Billiton to the model portfolio in 2011 because its combination of rapid growth and apparently low price made it an attractive option for my then-embryonic “defensive value” investment strategy.

With hindsight, BHP’s growth rate was completely unsustainable and the purchase price was too high rather than attractively low, so the final results are not exactly impressive.

However, it was a mistake worth making because it contained some important lessons for the future, which I’ll outline below. But first, let’s have a look at the final result.

Headline investment results: Lots of risk but not a lot of return

That table shows the final result, but it doesn’t show the volatile route BHP’s shares took to get there.

For that, we need to look at a share price chart:

And if you’re wondering why BHP would be so heavily affected by the price of oil, here’s an overview of its business:

BHP Billiton: A brief overview

BHP Billiton’s strategy is simple: It buys, develops and operates large, long–life, low-cost “upstream assets” such as coal mines, oil fields and iron quarries.

Those upstream assets produce commodities which BHP sells, but commodity prices are volatile and that can lead to volatile profits. To reduce volatility, BHP owns a wide range of assets which are diverse by both location and commodity.

BHP has been doing this since 1851. Back then it was called Broken Hill Proprietary, named after the location of its first mine in Broken Hill, Australia. When it joined the model portfolio some 160 years later, BHP was producing net profits of more than $23 billion.

So that’s the high-level overview, and it kind of begs the question: Why did I, as a self-confessed defensive value investor, buy into a company as volatile and highly cyclical as BHP Billiton?

Well, here’s the answer:

I bought BHP in 2011 because it combined dazzling growth with a low share price

Unfortunately, my notes on why I bought BHP are minimalist, to say the least. In mid-2011 there was no UK Value Investor newsletter, so I didn’t write anything in there. I also didn’t write a blog post about it, which is odd.

What I do know is that my strategy at the time was heavily based on a book by Mary Buffett called The New Buffettology. Even today there’s a lot of overlap between our approaches; mostly our focus on companies with ten years of unbroken dividend payments and steady growth.

But back then my strategy was much closer to the Buffettology approach than it is today.

So although I don’t have a written record of my thoughts on BHP at the time, I do know it fit the Buffettology criteria of having good growth, good growth consistency, a strong balance sheet and a low price relative to past dividends and earnings.

In fact, it’s easy enough to get hold of BHP’s financials up to 2011 from a data provider like SharePad, plug them into a suitable spreadsheet and see how the company would have looked through the lens of the metrics I use today.

Here’s a chart of that data, which clearly shows why BHP was an interesting prospect at the time:

That near-vertical increase in profits following the financial crisis gave BHP an incredible 26% annual growth rate over the previous decade.

It’s growth quality score (i.e. consistency) was an impressive 92% and yet BHP’s shares had a dividend yield of 3.3%, not far below the market average at the time.

Its PE10 ratio (price to ten-year average earnings) was also relatively benign at 16.7, which again was very close to the market average.

So BHP was a market-leading company with an extremely high growth rate which the market was valuing as if it were no better than the average company.

That, in a nutshell, is why it joined the model portfolio. Of course, I realised that a 26% growth rate wasn’t sustainable, but I had no idea what sort of growth might be possible.

The Buffettology strategy used a system of calculating possible future returns based on the amount of profit reinvested in the business and the expected return on that investment, but that gave an estimated long-term growth rate of 20%, which was optimistic to say the least.

What was missing from my analysis in 2011 was a real appreciation of just how cyclical commodity prices and commodity-related companies can be.

And thanks to BHP, I soon gained the appreciation I had been missing.

Holding BHP as the commodity super-cycle proved worthy of its name

The assumption underpinning both the Buffettology and defensive value investment strategies is that the past is a reasonable guide to the future. It’s far from perfect, but it is at least usable, most of the time.

So if a company has a long track record of consistent profitable growth, the default assumption is that there’s a reasonable chance growth can continue. And most of the time I think that’s true. But it isn’t always.

One example of unsustainable growth is where growth has been driven by increasing amounts of debt:

• Money is borrowed from the banks and used to fund expansion, factories, machinery and so on.
• This drives growth, but the increasing debt eventually makes the company extremely vulnerable to profit declines.
• In a downturn, profits fall but interest payments don’t, so those interest payments cannot be met.
• The banks can take control and the company’s high growth story has a very unhappy ending.

Another example is where rapid growth is driven by high demand, high prices and supply which takes years to catch up with demand. For example:

• The price of oil goes to $200 per barrel.
• Oil producers who can suck oil out of the ground for $50 per barrel make incredible profits.
• Those profits are reinvested into new oil fields and new supplies.
• When that new supply comes on stream it fails to meet demand, so prices remain high.
• As long as demand exceeds supply, existing and new upstream assets continue to generate huge profits.
• Because prices remain high, more and more supply is brought online as everybody tries to get in on the action.
• It may take years, but eventually, supply exceeds demand.
• Prices collapse, and many of the huge investments in supply are now generating massive losses at these lower prices.
• For physical or psychological reasons it’s often hard to turn the supply off, so prices stay low and assets continue to generate losses for years.

That, in a nutshell, is the commodity cycle. More specifically, it’s the China-driven commodity super-cycle which went from super-boom to super-bust somewhere between 2011 and 2014, depending on which commodity you’re looking at.

Despite BHP owning a diverse range of upstream assets, the company’s financial results and its share price were hammered as the price of just about every commodity took a synchronised dive.

For a few years, the damage seemed to be contained and the dividend was sustained, but when the oil price collapsed from $130 per barrel in 2014 to $30 per barrel in 2016, BHP had no reasonable alternative but to ditch its progressive dividend policy and cut its dividend.

So things didn’t turn out as I’d hoped, and that’s fine as long as a) I understand what went wrong and b) I change my investment strategy to avoid repeating the mistake.

And both of those points relate to why I sold BHP a few days ago.

Selling BHP because its share price no longer offsets the enormous uncertainties

I’m a firm believer in the idea that most of what happens to companies on a day-to-day basis is unpredictable noise and that in the long-run only a handful of factors really matter.

For BHP, the deciding factor in whether it succeeds or suffers is commodity prices.

Unfortunately, commodity prices are volatile and highly unpredictable, which means that BHP’s future revenues, profits and dividend payments are also volatile and highly unpredictable.

This undermines the assumption I mentioned earlier, the one that says past growth is a reasonable indicator of possible future growth.

If a company like BHP can go through ten years of boom, followed by ten years of bust, then past growth is not a good indicator of future growth.

At this point, I could just decide to exclude commodity-related sectors from the model portfolio. However, I still think these companies are investable, especially market leaders like BHP.

The trick is to remember that commodity-related companies are highly cyclical and that cycles have booms and busts.

Where I went wrong with BHP was to buy the company at a price that seemed reasonable as long as its high growth track record translated into at least half-decent future growth.

But it didn’t, and it was never going to because that growth was based on a cyclical boom, and cyclical booms are by their very nature unsustainable.

What I should have done was assume that BHP’s past growth was not likely to continue.

I should have assumed that its future growth would be average at best; perhaps 5% per year rather than 20% or more.

Based on that assumption its share price would have to be much lower before it looked attractive, as a lower price and higher dividend yield would be required to offset the lower expected growth rate.

I’ve been through this before with Braemar Shipping, Rio Tinto and BP, all of which joined the portfolio around 2011 and all of which generated relatively poor annualised returns.

The lesson I learned from those investments was to have a much stricter limit on the price I’d be willing to pay for highly cyclical companies. Specifically:

• I now insist on PE10 and PD10 ratios of less than 10 and 20 respectively for highly cyclical companies, rather than the usual 30 and 60 for defensive and “normally” cyclical companies.

If I’d used that rule in 2011 it would have kept BHP out of the model portfolio until the share price was much lower.

There’s also a significant possibility that BHP wouldn’t have made it into the portfolio at all.

And while I can’t be sure that the model portfolio’s performance would have been better with that tighter valuation rule in place than without, I do know it would have reduced the drag that commodity stocks have had on the portfolio over the last few years.

And that’s why BHP was a mistake worth making.

Finally, as usual, the proceeds from the sale will be reinvested into a new holding next month.