Sunday, August 29, 2010

Petra foods

Consists of 2 businesses:
  • Branded consumer division, which manufactures their own branded confectionery products and distributes 3rd party products in SEA (mainly Indonesia).
  • Cocoa ingredients division, which processes cocoa into chocolate ingredients used by chocolate manufacturers (e.g.: Mars, Cadbury, Meiji)
The Consumer division is well established, the Cocoa Ingredients is still starting up. In 09, Cocoa Ingredients accounted for 70% of EBITDA, despite both divisions having similar revenue.

Cocoa Ingredients

Involves processing cocoa beans into cocoa powder, cocoa liquor and cocoa butter. The last two are commodity products, the last may have some value add (by customizing to different customers needs).

Business Model:

High capex, high volume, low margins. Borrow money. Build processing plant. Borrow more. Buy large quantities cocoa beans, process them, and sell the resulting commodity products at a slight profit. The low margin nature of the business is shown by single digit margins for EBITDA (No Gross & Nett margins were provided by business segment).

Revenue (U$ mil)
Margin (for EBITDA)

They use derivatives to hedge against cocoa price movements. Management says they consider "the cocoa market to be a forum for managing risk rather than an opportunity for profit." [p5 09AR] i.e.: no trading profits. Gross cocoa grinding margins are not provided by Petra and its competitors for comparison.

Compeditive Advantage:

[From Apr 2010 S&P report, SGX research scheme]
One of the largest cocoa grinders in the world, based on grinding capacity:
  • 14%: ADM - publicly listed.
  • 14%: Cargill - private
  • 12%: Barry Callebaut - publicly listed, vertical integrated chocolate provider.
  • 11%: Petra
  • 5%: Blommer
Petra is large enough to stay around... but in this sort of market, the top players are evenly matched: everyone is a price taker.

Its top three customers, Nestlé, Cadbury and the Mars Group each account for about 5%-7% of revenue.


Cocoa grinding margin is affected by the difference in the price of unprocessed cocoa beans and the finished product. This seems cyclical - there was a glut of grinding capacity in 2002, and there was some margin compression last year after ADM and Cargill added capacity before the 2008 crisis.

(Wikipedia) If the combined butter and powder price is less than 3.2 times the bean price, grinding becomes uneconomical. Historical average is 3.5. See 'combined cocoa ratio' - grinding cocoa beans produces equal amounts of cocoa powder and cocoa butter (cocoa liquor is an intermediate produce, irrelevant here).

High cocoa bean prices per se do not affect their profit margins, as it is run on a cost-plus basis. Cocoa beans are bought (or secured through futures) after receiving an order. However record high prices increase working capital requirements, which hit Petra's cashflow in 2009.

The industry is highly cyclical, depending on the grinding capacity and the demand for it. I cannot predict the cycle, but Petra should be large enough to ride it out.

Industry Trends:
Continued increase in outsourcing trend expected. Attractive for chocolate makers to outsource the capital-intensive production of ingredients.

Branded Consumer

Business Model:
Manufacture and distribute their products (48% 09 revenue). Distribute 3rd party products (52% 09 revenue).

Competitive Advantage:

Petra has a more than 50% share in Indonesia's confectionery market. Closest competitor is PT Mayora Indah’s Beng-Beng (14.5% market share). Combined mkt share of MNCs is less than 10%.

Large distribution network, with 110 air-conditioned stock points and a fleet of delivery vehicles delivering directly to 70,000 outlets.


May be affected by:
  • cocoa, sugar, milk price.
  • general economy. Branded Consumer revenue dropped 11% yoy in 4Q08, after rising
Whole Company

Free Cashflow

Petra has been consuming cash for the last 6 years:


Notes on CFI:
  • 07: 28m PPE (15.8m brazil + europe, and 12.1m upgrade indonesia production capacity). Remainder 22m: acquire 70% Hamburg plant
  • 08: Almost all PPE: 76m to cocoa processing. 59m of that to Hamburg.
  • 09: Half PPE: 41.3m for cocoa processing. 26m of that to Hamburg. Hamburg finished.
  • 1H10: 70% is for purchase remainder of Hamburg plant.
Capex has dropped significantly in 2010 - management stated that have stopped all 'non-critical' capex.

In 09 and 1H10 CFO ballooned due to inventories. Inventories rose due to:
  • rising cocoa prices (25% in 2009):

  • Increased volume:

Cocoa Ingredient division - Volumes processed (000's mt)
119 (up 12% yoy)

Balance Sheet

High debt. Long term debt may be 5 to 10 times 2010 earnings. And it may increase more.

Total debt at 1H10 is U$514m. Of which 314m is current and 200m is long term. For comparison, FY09 earnings were 19m, earnings for 1H10 (alone) were 17.8m

Three things to watch:

1) Inventories. During 1H10, Cocoa inventories (for Cocoa Ingredients) rose 16%, while Cocoa processing volume rose only 12% yoy (search for 'mt' in Financial Results). Cocoa prices fell over this period:


Why are inventories rising faster than production, when the price of cocoa has fallen? It may make sense as they are ramping up production, and I do not know the exact time the inventories were bought. Not a red flag yet, but something to watch for.

2) Debt Maturity

Most of the 314m current debt is funded by short term borrowings:

For the U$200m non-current debt, $US130m of it is due between 2011 and 2014. They will have to refinance in the next few years:
  • 45m MTN due between 2011 and 2014
  • 25m MTN due between 2011 and 2013
  • 20m Term loan due 2011
  • SGD 60m (approx U$45m) MTN due 2012
Management said they are starting to fund inventories using medium term funding, I have not seen this appear in the numbers yet.

3) When can they start generating free cashflow to pay off their debt? Thev'e stopped capex, any remaining cost is working capital.

A simple back-of-the-envelope calculation... As most WC is for inventory, this would depend on their processing volume and the cocoa price. Assume their inventory level in 1H10 at 400m was for the 87% utilization that period. Assume they increase to 98% utilization, then another 10% is needed: 40m at current prices. If cocoa prices change, this affects the all the inventory (i.e.: the 400m, not just the 40m):

Worst case, if Cocoa prices go up another 100%, they may need U$500m. Best case if they drop 50% then reduced by 180m.
This stock is very illiquid. Mkt Cap @ $1.00 is 525m. Approx 82% held by insiders (those with > 5%), gives free float of 105m. Fund managers cannot buy. Not trade-able, even by retail investors. Suitable for long term fundamental buy only.

Indonesian family owned business.


Consumer business is a cash cow, with high barriers to entry. I like it.

Ingredients processing is a cyclical, commodity business, with high capex and working capital requirements. I don't like it. They have spent heavily since listing in 03 to become one of the largest cocoa processors in the world (processing capacity up from 7% (2003) to 14% (2010) of worldwide capacity). However they will never gain pricing power in such a fragmented market which has several strong players. Capital expenditure is now over after 1Q10, but more working capital may be required if cocoa prices rise (40m to 500m).

From comparing the two divisions, we can see why the chocolate makers are outsourcing their production....I wonder why Petra is doing the opposite? Well... if they didn't raise money to do this then they would never have listed in the first place...and we wouldn't have a chance to buy in.

Wait to see:
  • Inventories vs utilization vs cocoa prices
  • When can they start generating free cashflow?
  • Then do they pay off their debt?
  • How do they refinance their remaining debt?
And lastly... wait a while, years if need be... for the market to drop...

Saturday, August 28, 2010

Neither here nor there...

The US market's uptrend came under pressure 11th Aug, correction on 24th Aug. To me, this 'uptrend' was not significant enough to position trade. IBDs ratings do not work in a trendless market. I do not yet have this crucial skill to judge the state of the market (up, down, trendless) without hindsight.

However, following IBD does work well in a bull market. The previous examples I saw were from bull of 2003-2004. Waiting for a FTD, when the (brief) corrections are over, buying stocks that did not fall would have been profitable then. I'll keep my subscription to IBD while I see if I can learn to judge the markets better.

I have done nothing since June. Zero gain, zero loss. Right now, I do not know if the current market is still in a trading range, or correcting.

I am not cut out for swing trading, especially with a full time job. Position trading is difficult as I can't judge the market. I am changing my strategy to wait for stocks to be cheap, similar to end-08 or the 2001-2002 bear market. I may have to wait another 5 years....

Resist the urge to do something....

[Update: 13th Sep 10]
There was a FTD on 1st Sept. I am still doing nothing.