Sunday 29 May 2016

Prefs and Perps are Generally Inferior to Stocks and Bonds as an Investment Form

I have been investing in preference shares (prefs) for many years, so much so that I had forgotten that as an investment form, prefs are actually inferior to stocks and bonds. In fact, prefs have been the cornerstone of my equity-centric investment strategy, saving my equity portfolio during severe market crashes, as described in Behind Every Successful Bear Market Recovery is A Cash-Like Instrument. It was only last week when Hyflux launched its IPO of perpetual capital securities (perps) that I looked into the features of perps and remembered that prefs and perps are inferior to stocks and bonds as an investment form. 

Just to recap the characteristics of prefs and perps, they are perpetual securities with no redemption dates and hence are considered some form of equity. Unlike stocks, they have a pre-determined dividend/ distribution rate (distribution shall include to mean dividend for prefs and interest payment for bonds henceforth). However, the actual payout of the distribution for prefs and perps is discretionary, which means that the company can choose not to pay the full amount of distribution at the pre-determined payout date. For cumulative prefs and perps, any distribution not paid will accumulate and be paid out in the future (again, at the discretion of the company). Unlike bonds, this deferral of distribution is not considered a default event.

Compared to bonds, prefs and perps are inferior for several reasons. Firstly, payout of the stated distribution rate is not guaranteed and is not considered a default event. Secondly, if a company misses a bond interest payment and is unable to remedy it within a given period of time, the entire bond issue becomes redeemable immediately. If the company is unable to redeem its bonds, the company will go into default. Not only that, the company might have some loans with cross-default covenants stating that if the company defaults on some other loans or bonds, that particular loan will also become repayable immediately. In other words, when one creditor descends on the company, all other creditors will also quickly do so to safeguard their own interests. Thus, the company will be very careful not to default on any of its loans and bonds. For prefs and perps, there are no similar chain reactions since deferral of distribution is not considered a default event. Thirdly, in the event of claims, bonds rank higher than prefs and perps. Thus, bondholders have priority claim over pref and perp holders. 

Thus, for the above reasons, prefs and perps are inferior as an investment form to bonds. To compensate for the higher risks of prefs and perps, the margin of safety for investing in them needs to be higher. If you refer to The Lost Art of Bond Investment, you will see that the margin of safety required for Benjamin Graham's 2 criteria of minimum average earnings coverage and minimum current stock value ratio is higher for prefs than for bonds.

Compared to stocks, prefs and perps are also inferior for several reasons. Firstly, the upside is limited to the stated distribution rate for prefs and perps but unlimited for stocks. In my opinion, this makes for an unfavourable upside/ downside ratio for prefs and perps. Consider a pref with a stated distribution rate of 5%. If nothing goes wrong, you get a maximum of 5% in distribution every year. But if something goes terribly wrong, you can lose up to 100% of your capital. For stocks, it is also possible to lose 100% of your capital, but the capital gains (if any) for such high-risk stocks are usually more than 5%. Secondly, stocks have voting rights, but prefs and perps do not. If the current management is not doing a good job, stockholders can band together and vote out the current management. Pref and perp holders do not have such rights. Thirdly, the trading liquidity for stocks is much higher than that for prefs and perps. If you do not think the company is doing a good job, you can just sell and go away. However, for prefs and perps, there might be insufficient liquidity to sell at a reasonable price quickly. 

Thus, the worst possible scenario for prefs and perps is that the company defers some parts of the stated distribution and the price declines to reflect the uncertainty in timeline of the distribution, but you cannot get out because the company has no obligations to redeem the issue and the liquidity is too low for selling at a reasonable price.

In conclusion, as an investment form, prefs and perps are generally inferior to stocks and bonds. However, individual prefs and perps may have their own investment merits. Investors need to assess for themselves whether individual prefs and perps are good investments. 

See related blog posts:

Sunday 22 May 2016

Do Hyflux's 6% Perps Have Sufficient Margin of Safety?

Hot on the heels of Aspial, Perennial and Oxley, Hyflux has launched an IPO for its retail 6% perpetual capital securities (perps). I used to own its existing 6% preference shares when it launched in 2011. However, that was before I calculated whether it had sufficient margin of safety based on Benjamin Graham's criteria. After I calculated the figures, I sold it off in Jan 2014. Hyflux's preference shares were the first bond / preference shares that I evaluated using Benjamin Graham's criteria of minimum average earnings coverage and minimum current stock value ratio as discussed in The Lost Art of Bond Investment. When I posted it in Jan 2014, I did not expect retail bonds to become so popular. The criteria have been very useful in assessing whether the various retail bonds launched since Aug last year have sufficient margin of safety.

Since the post 2 years ago, has Hyflux's financial strength improved and do its 6% perps have sufficient margin of safety? Based on Hyflux's latest financial statements, the computation of the earnings coverage and stock value ratio are as follow.

Earnings Coverage
Profit before tax = $38.8M
Adjusted for:
- Add: Share of losses from associates = $19.6M
- Add: Finance cost = $42.8M
Total earnings available for covering fixed charges = $101.2M

Current finance cost = $42.8M
Add: Dividend for preference shares = $24.0M
Add: Distribution for perpetual capital securities = $25.7M
Add: Distribution for proposed perpetual capital securities = 6.00% x $300M

= $18.0M
Total finance cost = $110.4M

Earning Coverage = $101.2M / $110.4M

= 0.92

The earnings coverage of 0.92 times is below the minimum average earnings coverage of 2 times for public utilities or 4 times for industrial companies.

Stock Value Ratio

No. of shares = 785.3M
Share price = $0.545
Market value of shares = $428.0M

Current borrowings = $1,423.9M
Current preference shares = $400.0M
Current perpetual capital securities = $475.0M
Proposed perpetual capital securities = $300.0M
Total bond value = $2,598.9M

Stock value ratio = $428.0M / $2,598.9M

= 0.16

The stock value ratio of 0.16 is lower than the minimum stock value ratio of 0.67 for public utilities or 1.5 for industrial companies.

Thus, based on the above figures, the proposed Hyflux's 6% perps do not pass both the earnings coverage and stock value ratio criteria. Hence, based on Benjamin Graham's criteria, the perps do not have sufficient margin of safety.

Sunday 15 May 2016

Understanding Saudi Arabia's Game Plan on Oil Price

Is the decline of oil price since Jun 2014 a cyclical decline or a structural change? This question has important implications, as it affects whether Oil & Gas (O&G) stocks, which have been severely beaten down, are bargains. If it is a cyclical decline, it is only a matter of time before oil price recovers to higher levels and good business returns to the O&G companies. However, if it is a structural change, the good old business might never come back to some of the O&G companies again. To understand whether oil price is in a cyclical or structural decline, it is important to understand the intentions of the largest oil producer -- Saudi Arabia, whose actions affect oil price and the fortunes of the O&G industry to a great extent. Please note that I am not an expert in O&G and this post is very much a speculation on Saudi Arabia's game plan on oil price.

For a long period of time prior to the oil crash in Jun 2014, Saudi Arabia and OPEC had been moderating oil price, increasing production when oil price was favourable and cutting back when oil price was declining. This worked well, so long as other oil producers could not take advantage of OPEC's cutback and increase their own production. Things changed with the shale oil revolution in the US, which increased US' oil production to 9.2 million barrels per day (mb/d) as at Mar 2016, nearly matching Saudi Arabia's production of 10.2 mb/d (source: Oil Market Report). The more OPEC cuts back on production, the more its rivals will step in to take market share from OPEC. In the long run, this is not sustainable as OPEC will steadily lose market share to other producers. Thus, in Nov 2014, Saudi Arabia and OPEC made an important decision not to cut production any more to support oil price. Oil price fell precipitously as a result. After 2 years of oil price decline, the lower oil price has started to take a toll on US shale production, reducing production by 0.6 mb/d from the peak of 5.5 mb/d in Mar 2015. 

The price war is not without cost to Saudi Arabia and other OPEC members. Fig. 1 below, extracted from a World Bank research note in Mar 2015, shows that OPEC producers derive a great proportion of revenue from oil production.

Fig. 1: Percentage of Total Revenue from Commodities

Fig. 2 below, from the same research note, shows that the fiscal break-even price of oil, i.e. the price at which government revenue evenly balances expenditure, can be as high as USD100 per barrel for many OPEC producers.

Fig. 2: Fiscal Break-even Price for Oil

At the current oil price of USD45, many OPEC producers have difficulties financing their public expenditure and need to dip into their past reserves. Thus, although it is in Saudi Arabia's long-term interests to sustain the price war and reduce competition, there are short-term transitional pains to be addressed. Hence, it is no surprise that Saudi Arabia and Russia agreed to conditionally freeze production in Feb 2016 and to discuss production freeezes with all major oil producers in Doha in Apr 2016. Although the Doha talks failed to reach an agreement, the impact is manageable as the world demand and supply for oil is projected to almost reach equilibrium in the second half of 2016 anyway (source: Oil Market Report).

To improve its ability to sustain a prolonged war price and protect its market share in the future, Saudi Arabia has made long-term plans to reduce its reliance on oil revenue and lock-in market share. The proposed listing of Saudi Aramco (Saudi Arabia's state oil company) is a step in diversifying its economy from oil exports (see Saudi Arabia Plans $2 Trillion Megafund for Post-Oil Era). Another key step to lock-in market share is its expansion of refining capacity. Already, Saudi Aramco has equity stakes in oil refineries worldwide with a refining capacity of 5.4 mb/d. These refineries provide a ready buyer for its crude oil. It plans to increase this to 8-10 mb/d (see Saudi Arabia rewrites its oil game with refining might). At 10 mb/d, it is nearly equal to Saudi Arabia's current oil production of 10.2 mb/d. When that happens, it no longer needs to fight for market share with other oil producers.

Coming back to the original question of whether oil price is in a cyclical decline or structural change. The answer is both. It is cyclical in the sense that it is only a matter of time before oil price recovers to higher levels, as projected by the Oil Market Report. Last week's post on The Demand and Supply for Oil also shows that the demand and supply for oil are both inelastic above a certain price. When both are inelastic, higher price is possible.

However, there is an important structural change to the cyclical nature of oil price in future: Saudi Arabia will maximise production regardless of the level of oil price and will not cut production to support oil price. It is akin to stepping full force on the accelerator pedal regardless of whether the car is going uphill or downhill. The net effect is oil price will become more volatile than in the past when Saudi Arabia and OPEC moderated oil price.

Although oil price changes are partly cyclical and partly structural, the effect on O&G companies is more structural. Firstly, when Saudi Arabia and other OPEC members produce to their full capacity, all other higher-cost producers will be left to fight for the remaining share of the oil market, profiting when oil price is high enough and making losses when oil price is low. O&G companies that provides equipment and services to the higher-cost producers will be similarly affected. Secondly, when oil price is volatile, companies will become more conservative in making investment decisions. Imagine that you are an executive deciding whether to place an order for an oil rig that costs USD 800 million and is delivered only 4 years later. You do not know whether the oil price then is going to be USD30 or USD80. How do you make that investment decision? Some business will return, but the good old roaring business is not going to return for some of the O&G companies. This is why I mentioned in my previous post there are important differences between oil price and the economics of O&G companies.

Just a reminder, I am no expert in O&G. I am only an investor trying to work my way out of the O&G mess in my portfolio. It is a battle that I have no confidence of winning.

Sunday 8 May 2016

The Demand and Supply for Oil

I have quite a number of Oil & Gas (O&G) stocks in my portfolio. Yet, I never go and find out how do the demand and supply curves for oil look like, until Jan this year. The demand and supply curves would provide a clear picture of the territory upon which the battle of O&G would be fought. I always thought that demand would be inelastic while supply would be elastic. It turned out that I was wrong. This is the model that I managed to find on the internet (source: Energy Matters).

Fig. 1: Demand and Supply for Oil

The various coloured dots represent the actual oil production in the world at different points in time. A curve is fitted through the dots to form the supply curve for oil. The demand curve for oil is conceptual only and is shown by the 2 sloping straight lines, representing the demand at 2 different points in time. Readers are strongly advised to read Energy Matters' blog post to understand how the supply curve was worked out.

As expected, the demand curve is quite inelastic, i.e. a large increase in the price of oil would only result in a small decrease in the quantity demanded for oil. The supply curve, however, has 2 parts. The first part has a fairly gradual slope. At this part, the supply for oil is quite elastic, i.e. a small increase in the price of oil would result in a large increase in the quantity supplied. The second part has a fairly steep slope, which means that the supply is quite inelastic, i.e. a large increase in the price of oil would only result in a small increase in the quantity supplied. The point of inflexion from the elastic part to the inelastic part is around USD35.

The demand and supply curves help to explain the historical movement in oil price.

Fig. 2: Historical Oil Price

As shown in Fig. 2 above, oil price has been very volatile since 2004, rising to as high as USD140 per barrel in 2008 and falling to as low as USD30 in 2009 and in Jan this year. If you refer to Fig. 1, most of the price action during this period took place above the inflexion point of USD35, where both demand and supply for oil are inelastic. When both demand and supply are inelastic, a small change in either the quantity of oil demanded or supplied would result in a large change in the price of oil, hence, explaining the high volatility of oil price during this period. It also explains why oil price could fall from USD110 in Jun 14 to USD28 in Jan 16 so quickly and by so much in the last 1.5 years.

Despite the current doom and gloom on oil price, there is a silver lining in Fig. 1. There is an inflexion point at around USD35, below which the supply for oil becomes elastic. When oil price continues to fall along this part of the curve, producers will have to cut back production by more as it is no longer profitable to operate some of the higher-cost oil fields. The more oil price falls, the more producers have to cut back, eventually establishing a new equilibrium.

When oil price fell to USD28 in Jan, there were talks that oil price could fall to as low as USD10! Is this possible? Yes, because in the short term, speculation and herd mentality could force oil price to any price. However, is this sustainable? Looking at Fig. 1, if oil price were to stay this low for a prolonged period of time, producers would have to cut back drastically. In the medium term, fundamentals will take over and lead oil price to a higher, more sustainable level. 

Currently, US shale oil producers are cutting back production, which is one of the reasons why oil price has recovered from USD28 to USD45. Would oil price go back to the good old days of USD80 and above? If you look at Fig. 1 again, both demand and supply are inelastic above the inflexion point, so it is not improbable. It might seem highly unlikely that oil price could return to such lofty levels again, but neither did many people believe that oil price could fall from USD110 in Jun 14 to USD28 in Jan 16. The demand and supply curves show that anything is possible. The model, assuming it is accurate, provides a map of the territory that oil price will move.

A word of caution: even if oil price were to recover to higher levels, it does not necessarily mean that good business will return to many of the O&G companies. There is a difference between oil price and the economics of O&G companies. Understanding this difference can mean the difference between survival and annihilation in the battle of O&G!

Another word of caution: I am no expert in O&G. I am only an investor trying to work my way out of the O&G mess in my portfolio. It is a battle that I have no confidence of winning.

See related blog posts:

Sunday 1 May 2016

Understanding the Accounting for Joint Ventures

Perennial has successfully launched its second, 4-year, 4.55% retail bond this week. For a discussion on whether the bond has sufficient margin of safety, you can refer to the analysis for its first retail bond here. The purpose of this post is to discuss the accounting treatment of joint ventures (JVs) and associates, using Perennial as an example as it has a fairly detailed breakdown of them.

Fig. 1 below shows Perennial's balance sheet as at Dec 2015, extracted from its annual report. As the info in the screenshots are rather small, please refer to the annual report for better clarity of the info.

Fig. 1: Perennial's Balance Sheet

As shown in the figure above, it has total current and non-current loans, borrowings and bonds of $2,055.6 million (shown in lines 12 & 13 in Fig. 1). Note that there is an item called "Associates and Joint Ventures" in the non-current assets that amounts to $1,975.1 million (line 6). Associates and JVs are generally consolidated in the balance sheet on a net asset basis. For example, assuming there is a 50-50 JV which has assets of $100 million and liabilities of $20 million, its net asset is $80 million. On the parent company's balance sheet, this JV will show up as an asset of 50% of $80 million, or $40 million. Its 50% share of the $20 million liabilities will not show up in the balance sheet. Likewise, neither will its 50% share of the $100 million assets show up in the balance sheet. The equity of the parent company is unaffected whichever way associates and JVs are accounted for.

Is it important to bring back the assets and liabilities of associates and JVs to the parent company's balance sheet? Yes, because it provides a clearer, look-through picture of the amount of debt that the parent company is carrying. For JVs, the rationale is quite clear, because the parent company has some control over the JV, similar to a subsidiary. For associates, however, it is quite subjective as the parent company may or may not have any control over it, depending on its shareholding in the associate.

Let us use Perennial as an example to understand how to bring back the borrowings of JVs and associates to the parent company's balance sheet. On page 218 of Perennial's annual report, there is a fairly comprehensive breakdown of the company's investments in JVs and associates and their respective assets and liabilities, as shown in the figures below.

Fig. 2: Perennial's Investments in JVs and Associates

Fig. 2 above provides an overview of Perennial's investments in JVs and associates. Besides equity investments, Perennial also makes shareholder loans to JVs and associates. Thus, when we discuss Perennial's share of borrowings in the JVs and associates later, we need to deduct its shareholder loans to avoid double-counting.

Fig. 3: Breakdown of JVs' Assets and Liabilities

Fig. 3 above shows that Perennial has 4 material JVs which have total current and non-current liabilities of $1,205.1 million. Not all of them are borrowings. In the footnote to the footnote, it states that the current and non-current financial liabilities are $609.7 million and $4.2 million respectively. Deducting Perennial's shareholder loans of $71.1 million, the net financial liabilities are $542.8 million. Perennial's shareholding in all these 4 JVs is 50%. Hence, Perennial's share of the JVs' borrowings is $271.4 million. 

Fig. 4: Breakdown of Associates' Assets and Liabilities

Fig. 4 above shows Perennial has 3 material associates, with shareholding ranging from 30% to 46.6%. There is no further footnote to indicate the amount of borrowings in these associates. In such cases, a reasonable assumption is to assume that non-current liabilities are mostly borrowings while current liabilities are mostly accounts payable. Perennial's own balance sheet in Fig. 1 shows this to be the case, where borrowings make up 93% of the non-current liabilites and 32% of the current liabilities. Using this assumption, the associates' borrowings are estimated to be $822.1 million. Deducting Perennial's shareholder loan of $74.8 million, the net associates' borrowings is $747.2 million. After accounting for Perennial's shareholding in the associates, its share of associates' borrowings is $233.1 million. If we apply the same assumption to the JVs' borrowings, Perennial's share of JVs' borrowings would be $130.9 million instead of the $271.4 million mentioned above.

We are now ready to bring back Perennial's share of borrowings in JVs and associates to its balance sheet. Perennial's borrowings shown in the balance sheet are $2,055.6 million as mentioned earlier. Its share of JVs' borrowings is either $130.9 million (based on non-current liabilities) or $271.4 million (based on footnote disclosure). Assuming that we want to be conservative and account for its share of associates' borrowings, another $233.1 million needs to be added. Thus, on a conservative basis, Perennial's look-through borrowings are between $2,419.6 million and $2,560.1 million. The look-through borrowings are $364 million to $504.5 million more than that shown in the balance sheet.

In conclusion, the usual accounting practice for JVs and associates is to show them as a single line item in the balance sheet on a net asset basis. The proportional share of assets and borrowings of JVs and associates do not show up. To have a clearer, look-through picture of the parent company's borrowings, it is necessary to look into the footnotes and adjust for the company's share of JVs' and associates' borrowings.

See related blog posts: