Tag Archives: PRIS-B

Portfolio transactions (EHL.AX, STPFQ, PRIS.A/B)

Emeco Group (EHL.AX)

I exited a couple of positions the past few days for various reasons. Today I sold my stake in Emeco Group because I think that my initial thesis is broken. I bought the company because I believed that the assets would be worth more than the current enterprise value of the business, and that the company would earn enough to manage their debt load.

Based on the latest trading update both now appear to be false. The company announced one week ago that it would close the Indonesian business segment and that they would incur a A$38.5 million impairment charge on their equipment. They have also relocated approximately A$10 million of their Indonesian fleet to Australia. Given the fact that their Indonesian fleet was just A$81 million at the end of this year this means that the impairment was approximately 54% of original book value. If you would apply the same discount to the remainder of their fleet it would completely wipe out equity.

So I think the margin of safety that I thought was there simply isn’t there. I also didn’t like that they had to refinance their debt at a high interest rate to get rid of some covenants. I originally thought they would be able to meet them by simply selling equipment.

Suntech Power Holdings (STPFQ)

I also exited my short in Suntech, and this is something I should have done a few months earlier because at the current low price of the stock the carrying costs of the position are a bit too much (despite the fact that I was synthetically short using a structured product instead of having to borrow shares). The short position returned 56.4% in a bit more than a year time, a result I’m happy with since the leverage of the position was less than one.

Prisa capital structure arbitrage (Class A/B shares)

When I wrote about the irrational pricing of the Class B shares compared to the Class A shares the B shares were cheaper than the A shares despite the fact that they were superior in almost every way. We are now nearing the automatic conversion of the Class B shares in the Class A shares, and I managed to exit most of my position at a fair premium. The automatic conversion will be at 1.33x A shares for every B shares, and there will be a 24.8% dividend on the B shares as well (payable in A shares).

Calculating the performance of a long/short trade isn’t easy, especially when you have traded a bit and when one leg is traded in euro’s and the other leg is traded in dollars. So the results below aren’t totally accurate. I have measured the gains and losses in the table below compared to the initial value of the short leg (the Class A shares):

Gross total return 130.5%
Borrow fees -12.0%
Trading costs -3.1%
Net return 115.5%
Return attributable to dividend payments 14.3%
Return attributable to premium increase 55.6%
Return attributable to residual long exposure 33.0%
Return attributable to trading profits 12.5%

I think this might be one of my best trades ever. That the outcome would be this favorable was certainly not a given: if Prisa would have entered bankruptcy I would probably have made just a little, or lost a little. But being able to double your money without significant market exposure is obviously pretty awesome :).


Still long some Prisa Class B shares, and short a few Prisa Class A shares. No position in STPFQ or EHL.AX anymore.

Pricing discrepancy in PRISA capital structure

Promotora de Informaciones (PRISA) describes itself as world’s leading Spanish and Portuguese-language media group. The company gets most of it’s revenue’s from Spain, but it also has a sizable presence in Latin America. It’s heavily indebted and not doing very well. But that doesn’t matter a lot for the following idea: even a bankruptcy shouldn’t be a disaster, although It’s not a best case scenario.

PRISA has two different share classes: ordinary Class A shares and convertible non-voting Class B shares. The B shares are convertible to A shares at any time. From the prospectus:

Holders of Prisa Class B convertible non-voting shares may at any time give Prisa notice of their election to convert the shares into one Prisa Class A ordinary share for each Prisa Class B convertible non-voting share. Prisa’s board (or a duly authorized committee) will, within five business days following the end of each month, issue Prisa Class A ordinary shares in respect of the Prisa Class B convertible non-voting shares whose holders have elected conversion during that prior month. Prisa will register with the Mercantile Register all Prisa Class A ordinary shares issued upon conversion as soon as practicably possible before the end of the month in which the Prisa Class A ordinary shares are issued.

So logic would dictate that the price of the Class B shares is at least equal to the price of the Class A shares. The market is currently failing to recognize this with the PRIS ADR trading at $1.70 at the time of writing while the PRIS-B ADR is trading at $1.57: that’s an eight percent discount for a security that has at least an equal value. Both ADR’s contain four shares.

While this is already noteworthy the discrepancy is even bigger when you realize that the Class B shares are actually far better than the Class A shares. The Class B shares not only receive a dividend (7 Class A shares for 40 Class B shares), they will also be automatically converted to Class A shares in 2014. When the Class A shares are trading below €2.00 the conversion will be increased from 1:1 to a maximum of 1.33 A shares for every B share:

Any Prisa Class B convertible non-voting share still outstanding on the date that is 42 months after its issue date will automatically convert into one Prisa Class A ordinary share, without any action by the holder. In the event of automatic conversion, if the volume-weighted average price of Prisa Class A ordinary shares on the Spanish Continuous Market Exchange (Sistema de Interconexión Bursátil-Mercado Continuo) during the 20 consecutive trading days immediately preceding the conversion date, or the twenty-day trailing average, is below €2.00, then the conversion rate will be modified. In this event, the number of Prisa Class A ordinary shares into which each Prisa Class B convertible non-voting share will convert will be equal to a fraction (expressed as a decimal), the numerator of which will be €2.00 and the denominator of which will be the twenty-day trailing average, subject to a maximum conversion rate of 1.33 Prisa Class A ordinary shares per Prisa Class B convertible non-voting share. If the twenty-day trailing average is less than €2.00, Prisa may also choose to retain the 1:1 conversion ratio, in which case Prisa would pay a per share amount of cash equal to the difference between €2.00 and the twenty-day trailing average, subject to a maximum of €0.50 per Prisa Class B convertible non-voting share. The balance of the premium reserve in respect of the Prisa Class B convertible non-voting shares, if any, will be made available to pay the nominal value of the Prisa Class A ordinary shares to be issued in excess of the Prisa Class B convertible non-voting shares to be converted.

With the Class A shares currently trading at €0.335 in Madrid it seems quite probable that holders of Class B shares will not only receive a few years of dividend payments before the automatic conversion in 2014, but that they will also receive 1.33x Class A shares for every Class B share bought today. So I think it’s quite obvious why I think the current market pricing is just wrong, and I haven’t even mentioned yet that the Class B shares rank above the Class A shares in case of a liquidation. While a liquidation is probably not a disaster for a long/short trade it’s not the best outcome if it would mean that both classes go to zero.

In a liquidation of Prisa, the Prisa Class B convertible non-voting shares would be entitled to receive, on a preferential basis according to applicable law, their stated value per share, before any distribution is made to the holders of Prisa Class A ordinary shares. In the event that Prisa has, immediately prior to any liquidation, distributable profits or share premium reserves in respect of the Prisa Class B convertible non-voting shares, the holders of the Prisa Class B convertible non-voting shares would receive any unpaid minimum dividend, including any accumulated unpaid dividends from prior years, in respect of the prior and then current fiscal year.

Holding the Class B shares to maturity and hedging the exposure to PRISA with a short Class A position is a pretty decent idea, but it’s also possible to buy PRIS-B shares and convert them directly to Class A shares. Not only will you capture the price difference between the two share classes, you also receive the accumulated dividends. Based on the declining number of Class B outstanding and the increasing amount of Class A shares there are people who are doing this trade. Last month 12 million Class B shares were converted to 14.1 million Class A shares (the dividend is paid in Class A shares).

Both trades have one big risk: being able to maintain the short position in the Class A shares as a hedge. There aren’t a lot of shares available for shorting, and you are paying at the moment 5% annually for the borrow. If you lose your borrow you could be forced to buy-in at exactly the wrong time, or the borrow fee might go through the roof. Think this risk is significant, but at the same time it’s really the only risk you are taking here.

A bit of good news is that the short term arbitrageurs should both increase the supply of Class A shares available for borrow and increase the price of the Class B shares. It’s also good news for the people who are long PRISA: a lower amount of Class B shares outstanding means less dilution in the future.


The PRISA Class B shares have been relatively cheap ever since they were created, and I have had a long position in the Class B shares since 2011 (my worst investment ever). But with the spread between the two share classes at an historic low, and the automatic conversion only getting closer I couldn’t resist buying some PRIS-B shares. This time I’m not making the mistake of not hedging my long exposure though.


Short PRIS, Long PRIS-B