Monday, December 3, 2012

Ships, hogs, dirts and the shipbuilder called Conrad

Ships, hogs, dirts and shipbuilders. What do they have in common?

Their economics.


A while back, the dry bulk shipping industry caught my attention. You can see why from the 3-year BDI index chart below. The BDI index is the barometer of the charter rates for dry bulks. The industry has been in recession since GFC.  It is severe. Besides, I remembered one of my role models, Mohnish Pabrai, described in his book how he scored a multi-bagger win in his bet on the oil shipping company Frontline when the industry was in recession in 2001.

(image: BDI Index)
Source: Bloomberg

Were there values among the dry bulk shippers? 

The initial look looked promising. Listed dry bulk operators like DSX and GNK spotted attractive ROAs, profit margins and P/B ratios. However, the more time I spent to understand the industry and its economics, the less sure this was a game I was capable to play.

Take a look at the supply/demand curve I reproduce from Martin Stopford's book Maritime Economics :

(image: supply/demand curves)
Source: Maritime Economics, by Martin Stopford

Let me point out the important bits:
  • Shipping is essentially a commodity business. (You generally don't care too much who is shipping you stuff as long as your goods arrives in one piece.)
  • In short term, demand is inelastic. (If you need to buy steel beams to build your Olympics stadiums, high shipping cost won't easily deter you.)
  • At the same time, worldwide shipping capacity is finite, because it takes years to build a new ship. Thus, supply becomes inelastic once you reach a certain point. Hence, the "hockey stick" shape supply curve. Freight rates can go from $6,000 to $44,000 in the space of a few months.
  • When freight rates skyrocket, shippers will decide to invest to expand their capacity. It takes 1-3 years to build a new ship. By the time the shipbuilders expand their shipyards and new ships are built, the demand is no longer there. We now have an oversupply of ships and freight rates tumble.
So, here we go. Boom and bust cycles. Because of the above structural reasons, the magnitude of boom and bust cycles is extreme. Consider this: BDI was at its highest 11,000 in mid-2008 before the GFC, tumbled to 700 in Dec 2008, recovered to 4,600 in Nov 2009 and is now hovering around 1,000. To get a sense of what this means in real life, just imagine bus fare tumbles from $110 to $7 and then climbed back to $40 in the space of months.

This means the usual metrics like ROA, P/E and P/B are all meaningless. Earnings and asset values are quick sand. They are extremely unstable. You can't rely on them to value dry bulk shippers.

I ended up not investing in any of them because I just had no particular insight into individual shippers. Nor had I any insight into the cycles and the macro environment surrounding them.

Hogs and Dirts

I've omitted a lot of details about maritime economics which contribute to the "hockey stick" supply curve. (e.g. Ship owners can slow down their voyage or lengthen their maintenance time in response to low demand.) But the above supply/demand captures the essence. Furthermore, there are 2 key factors underscoring this extreme economics: (1) The decisions to expand the supply (i.e. the fleet) take years to materalise. (2) Each player in the industry is making rational decisions, but only considers themselves in isolation. Some kind of prisoner's dilemma is at work here.

This pattern shows up in another industry that I'm been worried about for some time: the mining sector in Australia.

Professor Steve Keen at University of Western Sydney explained it the best in this Business Sepctator piece. He pointed out this is nothing new. This was long recognised in hog cycle, the volatile 4-year cyclical pattern of prices for pigs in the US. And there is a neat economic theory, the cobweb model, explaining it.

(This is a good example that knowledge is accumulative. You builds up your circle of competence organically over time. From time to time, Things I learned from one place would show up in another place in a slightly disguised form. Things learned from one domain are never wasted if they didn't lead to any investment idea.)

Conrad Industries

This brings us to CNRD, the shipbuilder that I'm investing in.

My original conservative estimation of CNRD's intrinsic value was $18-20 per share. The current share price has now fallen into this range. Isn't it time to take the money off the table? This is the question I've constantly had in my mind in the recent weeks.

The original investment thesis was essentially based on a single event, the oil spill, or the recovery from it. But I have since realised there is more with CNRD. CNRD's management is more competent than I initially thought. CNRD has also become less sensitive to the exploration activities in the Gulf region than it used to be as the management has diversified its client base. CNRD may not have any structural advantage, but it is a very efficient business. It has the appearance of a "hidden champion". It's more like Buffett's Nebraska Furniture Mart than his Coca-Cola.

The difficult question is: how to value it now?

CNRD isn't exactly Nebraska Furniture Mart. Even though CNRD's client base is more diversified now, the products it makes are still commodities. It is still at the mercy of boom and bust cycles. "When a management with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact." More importantly, the longer I hold CNRD, the more important the cycles will become.

For a cyclical business, the concept of intrinsic value as in Ben Graham's way of thinking may not even be applicable. Even if it exists, it's close to unknowable. There is no stable earning. It will be dangerous to normalise CNRD's earning over many years to arrive at an artificial figure. If we do so, we will be like having one foot in a bucket of ice water and another foot in a bucket of boiling water and claim we feel good on average.

And it is equally dangerous to time the cycles. Stopford says in his book the average length of boom and bust cycles in the shipping industries is about 7-8 years. This is, again, just an average. These cycles don't come in as clock work. The constant changing macro environment has significant influences on the supply-and-demand.

Where does this leave us? Without an estimation of the intrinsic value, I don't have a rational basis to judge when to sell. And forcing an estimation may easily give me a precisely wrong figure.

I'll follow Keynes' doctrine: "it's better to be vaguely right than precisely wrong". Here is how I will approach it:
  • I mentioned in a previous post there are a few catalysts surrounding CNRD: the BP settlement, full recovery of its repairing/maintenance segment and a small possibility of some kind of corporate action. So, instead of getting obsessed with figuring out a valuation, I will wait for one or more of events to play out.
  • I will add one more event to my list above. There is an important observation from the discussion of the boom and bust cycles of the shipping industries: The length of the cycles is a directly consequence of the duration it takes to materialise the investment/expansion decisions made by the players. It is not precise and will never be precise because unpredictable macro events will push things around. But the general cause and effect is there. If it takes only one week to add capacities in an industry, you will  expect to see the length of cycles in the order of weeks, not years. And here, we see CNRD is also buying land and getting government grants to expand its capacities. We can reasonably expect other players in the industry are doing the same within a similar time frame. (This will be a good place to do more scuttlebutt.) When all of these new capacities come online, we have to be worried. So, if we work backwards from here, the completion of CNRD's expansion program will be a signal the industry has passed its peak.
  • I will err on the side of being over-cautious. I will rather leaving too much money on the table than being caught in the downturn of the industry.
I mentioned in my previous post the Credit Bubble Stocks blog is an excellent source of intelligence on CNRD's industry. Credit Bubble has linked to an informative research piece Good Year for the Barges. Being a contrarian, I'm becoming more cautious, treading with my eyes wide eye...

(Disclosure: Long CNRD)