Friday 27 September 2013

Musings on Investor Psychology

After spending a few years in the investment game, I now have a firsthand appreciation for the difficultly in completely filtering out the impact that share price fluctuations and other market noise have on your decision making, which can also be inherently flawed on its own. More and more I realise that having an awareness of, and being able to control cognitive biases is just as important as understanding business. Many great investors have warned against the psychological challenges involved - Charlie Munger says: Above all, never fool yourself, and remember that you are the easiest person to fool - and I would second that. It would be very interesting to see some experts in psychology try their hands at investing. 

If you can spare 10 minutes or so, I would highly recommend you read this extensive list of cognitive biases on Wikipedia. While many are clearly applicable to investing, such as loss aversion and outcome bias, there are many other social and memory biases that it would be useful to be aware of for day-to-day life. But be careful that you don't pick up one or two extra biases in reading that list - 'bias blind spot' is 'the tendency to see oneself as less biased than other people, or to be able to identify more cognitive biases in others than in oneself.'  

As it relates to investing, misjudgements due to these biases can be very costly mistakes, which is a factor that is often overlooked. While nobody is completely immune to these cognitive biases, I believe they can be reduced by continuously scrutinising one's decisions and thought processes, and having an understanding of what the various types of biases are. On the latter front, I intend to read some books on psychology after my higher school certificate exams are over, which I expect will prove fascinating and rewarding. I think that the best thing about this blog is that it has forced me to clearly argue why I have done what I have, and since it is all there in black and white, I can look back to assess why I bought or sold something without any fear of memory bias. Both of these have helped expose psychological biases in my own thinking. 

Perhaps the stock that I have had to be the most careful with in terms of cognitive biases is Delta SBD (ASX:DSB), an underground coal mining services company I bought in late February this year. While I'll try to outline my thinking, I must first warn you that my judgement of the situation is likely to be quite biased since I still hold shares in it, so as always, I encourage you to be critical of my thinking.

I initially bought in DSB late February this year at $0.76, however it has since dropped to a low of $0.33 a couple of months ago, and is now back up to $0.53. Incidentally, the day I bought was the very peak, and ever since it has been a rather painful trip down. I knew going in that this was a fairly mediocre business (current normalised return on equity of 14%, net debt to equity of 32%, and a net profit margin of 5.5%), and anticipated a less than impressive earnings outlook, but the price looked cheap enough to offer a significant margin of safety - less than 5 times FY2013 earnings. In addition, I liked that the founders of SBD and Delta were still managing the company, and held a majority stake.

Unfortunately, I underestimated the speed and extent of the downturn in the coal sector, with news day after day reporting hundreds or thousands of job losses due to declining coal prices that made many mines uneconomical. Perhaps the widespread and sudden change in sentiment towards the sector is explained by the 'availability cascade', which is 'a self-reinforcing process in which a collective belief gains more and more plausibility through its increasing repetition in public discourse'. 

I will admit that in response to this outlook, I probably fell into the trap of confirmation bias - tending to read more articles that had positive views on China and the coal sector in an attempt to justify my initial investment. After losing 31.5% in a couple of months, I decided to double my initial investment by purchasing more shares at $0.52 since I believed that the market was being unnecessarily pessimistic and thus DSB had become cheaper. Eventually I recognised my confirmation bias mistake and started listening more to the bearish arguments (although I still have a healthy scepticism for any macroeconomic forecasts). In combination with the falling share price, it soon became hard to decide on what course of action to take - should I sell, buy more or wait for more information in the annual report? My value investing framework told me that despite the poor outlook, a share price in the 30 cent region was most likely too low so I ruled out selling at that price. However, neither did I have enough confidence to buy a third time, so in the end I opted to wait more information from the company, which had been deafeningly silent in its announcements.

Eventually, the annual report came which was largely as I expected, with a great FY2013 but with an expected drop in revenues and compression of profit margin in 2014 onwards. While there is now considerable uncertainty over the future earnings of DSB, they certain to be substantially lower, with a greater than 50% drop in earnings next year looking quite likely to me. That would take DSB from a bargain basement P/E of 3.1 currently to a still quite cheap P/E of 6-7, but there is considerable downside risk to that if China well and truly blows up sometime soon, which is a risk that makes me uncomfortable holding DSB for the long term.

To break-even on my two parcels, I would need a share price of around $0.59 (including fully franked dividends and brokerage). While it is tempting to hope to sell at a break-even price, this is another psychological pitfall known as 'anchoring' that inhibits rational decision making. As Phillip Fisher put it: More money has probably been lost by investors holding a stock they really did not want until they could "at least come out even" than from any other single reason. If to these actual losses are added the profits that might have been made through the proper reinvestment of these funds if such reinvestment had been made when the mistake was first realised, the cost of self-indulgence becomes truly tremendous. Instead of calculating whether you will come out even, what needs to be asked is - 'where is current share price in relation to its intrinsic value?' If the answer is 'lower' then it usually makes sense to hold, unless as Fisher reminds us, you have found a better opportunity that you require funds for. While I am aware that my judgement may still be clouded somewhat by owning shares in DSB, for now my valuation work suggests the answer is 'a little lower' and so I continue to hold for the time being. 

The verdict is still out on what the future of DSB, the coal sector and China is, but regardless of the outcome, I hope that I will have at least taken away some valuable lessons about investor psychology that will improve my decision making going forward.

Tuesday 3 September 2013

Antares Energy

Amidst the barrage of results in reporting season, I think one announcement has gone a little unappreciated. That was the news that Antares Energy (AZZ) had moved from a Letter of Intent to a binding Purchase and Sale Agreement to sell all of their Permian oil and gas assets for $300 million USD (for this post, I'll use the current exchange rate of 1 AUD = 0.903 USD, so that's $332.1 million AUD). I couldn't believe my eyes when I saw that the market capitalisation was $131 million, and this price had fallen over 6% on that day. Well it turns out that I was right not to trust my eyes, as further inspection reduced the attractiveness of this investment, but not enough to hold me back from purchasing.

First, a little background to AZZ. Back in around 2007, the management had almost sent the US focused oil and gas company bankrupt as their 'three year Strategic Plan' failed, recording a loss of $37 million, negative equity of 5 million, and negative cash flow. Shortly thereafter, the share price fell from over $1.00 to below 4 cents, and a shakeup of management occurred. A reversal in fortunes occurred as debt was brought under control and the operational side improved, with the share price rocketing back upwards to over 80 cents at one point. In FY2010, they announced that Chesapeake Energy was going to buy out the Yellow Rose and Bluebonnet assets that Antares Energy held for $200 million USD, of which $156.2 million USD would go to Antares for their proportional ownership. Further bolstering financials, 50.5 million shares were issued at an average price of $0.62, and management soon bought back 20.5 million shares at an average price of below $0.40. This action essentially netted them a profit of $4.5 million on the shares they bought back, which is a big tick in my book. And they've continued to do this, repurchasing 43.3 million shares in total for an average price of $0.413.

With a newly found cash pile, AZZ spent around $160 million on three projects in the Permian basin of West Texas in 2011 which are shale plays, and as you may know, the United States has benefitted from a shale gas boom in recent years. Reserves have grown significantly since then and this brings us to the present, where AZZ is expected to sell these assets on or before 15th January 2014 to an unnamed company (for commercially sensitive reasons). How much of the recent success is attributable to the shale gas boom and how much of it is due to good management I'm not sure, but either way, I give credit to their foresight in buying into the Permian basin when they did. From a terrible position in 2007 to being offered $332.1 million is quite an achievement.

Now, that $332.1 million isn't entirely going to be in the hands of shareholders. To begin with, AZZ currently has debt of $41 million USD and according to their most recent quarterly report, cash of $7.5 million AUD, giving net debt of $37.9 million AUD. Upon sale, the entire debt facility will be reduced to zero, leaving the only other major liability being the convertible notes. There are 10 million convertible notes on issue which have a face value of $2.00, with each note being convertible into 3 shares - i.e. it becomes economical to convert into equity at a share price above $0.66. If none of these notes convert into shares, that's an extra $20 million of debt that will have to be paid eventually (total conversion adds 30 million shares to the existing 255 million, having a similarly negative impact through dilution of shareholder interests). Although some brokers have been trying to guess about the tax consequences of the sale, throwing around figures of $25 or $35 million or even $60 million, I spoke with the chairman and CEO James Cruickshack over the phone, who surprisingly indicated to me that he expects Antares to pay zero tax from the sale. This is because Section 1031 United States Internal Revenue code allows businesses like Antares to dodge the capital gains tax if they purchase another asset within 180 days of the sale. Even if they don't purchase something else within that timeframe, Antares has until December 31 2014 to utilise any production/exploration costs as a tax deduction, so that in either scenario, no tax will be payable. And finally, there are the transaction costs, which I'll take as an extra $15 million (Mr Cruickshank mentioned the difficulty in determining the transaction costs, but confirmed that this number was in the ballpark range).

So there are a few scenarios which could happen:

1. The deal goes ahead and AZZ finds another asset in the Permian region or elsewhere to purchase (no convertible notes change into equity). Their net cash would be 332.1-37.9-20-15 = $259.2 million, or with 255 million shares on issue, net cash of $1.02 per share, double the current share price of $0.515. Should AZZ trade at this price when the sale is completed, a doubling of your money in 4-5 months is nothing to sneeze at.

2. Same as scenario 1 but all notes are converted into equity instead. Net cash would be $20 million higher as this convertible note liability would be gone, but shares on issue would be 285 million, giving net cash of $0.98 per share. (note: if only some of the notes are converted, the net cash per share will lie between these first two scenarios).

3. The deal goes ahead, and the entirety of the proceeds from the sale are returned to shareholders as a capital return. I consider this scenario rather unlikely given that management has previously bought more oil and gas assets with their proceeds, extended the maturity date on the convertible notes to October 2023, and after all, who voluntarily puts themselves out of a job? A more likely situation is the continued buyback of shares if they are cheap. (see *)

4. The deal falls through, perhaps due to shareholders not voting in favour of the sale, although I view this cause as unlikely. Another reason may be that the secret counterparty won't be able to pay, although the Board of Antares is confident that "they have the financial capacity and desire to complete the transaction". They also remind investors that they have a "100% perfect history in closing all transactions announced to the market". If for whatever reason it does fall through, the share price may drop substantially, but I am speculating that the market won't continue to assign such a low valuation to AZZ now that someone has offered them $332 million for their assets. There is also a purchase price adjustment clause in the agreement which says that if an adjustment is to be made to the purchase price if there is a 'Defect' found (eg. non compliance with environmental regulations), it will be no more than 10% of the purchase price. This means there is a $33.2 million downside to the figures I have outlined above, or $0.13 per share.

I would be quite happy with any of the first three scenarios, but there are some extra risks involved. Most obvious is the possibility of an adverse change in the AUD/USD exchange rate. If the exchange rate were to reach parity again, the net cash to AZZ would be $216.5 million AUD under scenario 1, or $0.85 per share. On the other hand, if the Aussie dollar continues to depreciate against the US dollar, net cash in Australian dollars would increase.

The second risk entails a bit of speculation on what the market will do in the short/medium term, which is something I try to avoid. Will the market value AZZ at precisely the net cash figures I have calculated? I went back to the last time Antares sold their assets and found that it traded at almost exactly its net cash per share. However, one cannot assume that this will be the case again, with the likelihood being that a discount will be placed on the valuation of AZZ since it will probably purchase another asset rather than pay out the cash to shareholders. In any case, I don't believe for a second that the market should value AZZ at just $0.515 and I expect that this mispricing will be corrected as January 2014 arrives.

The final risk is that the new acquisition/s will be a bad one, as acquisitions so often tend to be. I think this risk is somewhat mitigated by the fact that management are playing from a position of strength, not weakness (for example due to debt), in selling their Permian assets, and therefore are able to be opportunistic with their purchase. Who knows, the reason for this sale may have been because they have already identified a better opportunity. And as I highlighted above, current management have shown they are able to add considerable value through the acquisition of the Permian assets, so perhaps the market should in fact be placing a premium on the value of their net cash. Furthermore, directors hold 14 million shares, giving them an incentive not to squander shareholder money, and impressively, no director has sold a share since 2004. Nevertheless, an overpriced acquisition is certainly possible. I won't be around to see the next oil and gas play succeed or fail as I view this sector as out of my circle of competence, but I feel that I know enough in this situation to feel fairly confident about a short/medium term investment which relies on the market recognising fair value soon.

While this may not be a typical long term value investment for me due to the shorter time horizon and valuation based on net cash rather than future earnings, I still feel comfortable with this purchase given the considerable margin of safety provided by the cash in the bank. In the end, it is quite simple: if someone offered you a bank account with $259 in it, would you refuse the opportunity to take it off their hands for $131?

* Mr Cruickshank expressed his preference to me for continuing to buying back shares over a capital return or paying dividends as the easiest and most effective way to return money to shareholders, as long as the net tangible assets of the company exceed its share price and there are sufficient sellers on the market to repurchase shares. I agree with this approach taken, and see it as a sign of good management. To illustrate, lets assume a business has a share price of $1.00, 100 million shares outstanding, and its sole asset was net cash of $200 million, giving net cash per share of $2.00 (not dissimilar to the share price/net cash disparity that exists for AZZ). If management were to buy back 20 million of those shares at $1.00, that would reduce shares outstanding to 80 million, net cash to $180 million, and therefore increase net cash per share to $2.25. This really is buying $1.00 bills (or coins in the case of Australia) for 50 cents. The converse is also true: if shares were being bought back at a share price higher than the net cash, value would be destroyed. This gives me further confidence that management won't do something exceedingly silly with the funds they will receive. You might expect these principles to be obvious to CEOs and directors, but all too often I see shares being bought back at any price, without a justification as to whether the shares are undervalued or not.

** After giving further thought to the possible downside risk, and concluding there was little to no risk of significant capital loss, I subsequently bought an additional 1943 shares in AZZ at $0.515 on the 6th of September, 2013.