Monday, 29 January 2018

Market efficiency and Telecom Italia Savings Shares

The market is supposed to be efficient. There are a lot of smart and hungry investors out there competing vigorously with one another to feast over whatever bargains the market happens to be offering up, if any. These efforts are supposed to neutralise each other, and drive a high level of market efficiency. And yet, in practice, I continue to be confronted on an almost daily basis with the most bizarrely obvious mispricings, and sometimes to a degree I can scarcely believe.

A recent example was when I stumbled onto Telecom Italia's saving shares class earlier today. Telecom Italia is Italy's incumbent telecom operator, and I think the ordinary shares are mildly attractive at current levels. However, that is actually beside the point of this piece.* What is interesting is how the savings shares trade relative to the ordinary shares.

Telecom Italia has two classes of shares - ordinary shares and 'savings shares'. There are approximately 15.2bn ordinary shares outstanding, and approximately 6.0bn savings shares, for a combined total of circa 21.2bn. The ordinary shares have voting rights, and the savings shares do not, but for most minority holders, this difference holds little practical significance. The more liquid ordinary shares currently trade at 71c, and the somewhat less liquid savings shares trade at 61c, an approximate 15% discount.

In and of itself, a 15% discount for a less liquid class of shares that lack voting rights is far from remarkable or uncommon. Indeed, that is a fairly typical discount. However, where it gets interesting is when you take a closer look at the terms of these savings shares, which are highly atypical. They are outlined below in a snapshot drawn from page 346 of Telecom Italia's 2016 annual report:

In plain English, what this means is that the savings shares are entitled to an annual dividend of 2.75c per year,** and the ordinary shares are not entitled to this dividend. Furthermore, the savings shares are also entitled - on top of this 2.75c - to an additional dividend equivalent to any dividend paid on the ordinary shares. Currently there is no ordinary dividend, so the savings shares yield 4.5% (2.75/61), and the ordinary shares yield nothing. However, if the ordinary shares were to ever begin to pay dividends in the future, and declare - say - a 3.00c dividend, the savings shares would then be entitled to a dividend of 5.75c (the ordinary share's 3.00c dividend, plus their unique entitlement of 2.75c). Furthermore, the provisions highlight that the savings shares' economic rights would not be disadvantaged in any way in the event of a company wind-up/liquidation.

What this means is that the savings shares are materially superior to the ordinary shares, and yet they trade at a 15% discount. One way to conceptualise the savings shares would be to think of them as being akin to a stapled security, that offers (1) a quasi-perpetual bond or preferred stock yielding 4.5%; attached to (2) an ordinary share of Telecom Italia, which currently has a market value of 71c.

That the savings shares trade at a material discount to the ordinary shares is therefore a patent absurdity, even after allowing for some discount for the lack of voting rights. We are currently living in a world with exceptionally low interest rates, which is resulting in investors everywhere paying silly prices for any sort of yield they can get their hands on. Long term junk bonds in Europe can now yield as little as 2-3%, and Telecom Italia is a profitable (but still bloated) business at the bottom of the economic cycle with management that is working to improve operations. If this dividend entitlement was spun out into a separate class of preferred shares or covenant-light perpetual bonds, I am willing to bet that it would trade at more than 61c all by itself. And yet this quasi preferred yield security also comes attached with a full ordinary share that is currently changing hands in the market for 71c!

This is a remarkable mispricing. The only way I can explain it is that investors are habituated to seeing less-liquid preferred classes of securities lacking voting rights trading at discounts to the ordinary shares of approximately 10-20%. Consequently, investors may have merely assumed that 15% is the right discount level, without bothered to scrutinize the actual terms of these securities (or at least, thinking very hard about them).

It doesn't say a lot about the level of due diligence being undertaken by investors, and it stands as a remarkable testament the degree of market inefficiency that continues to exist in the real world, outside the hallowed halls of academia. This sort of obvious mispricing should not exist, and a one man generalist band like myself looking at dozens of securities a day, half way around the world, should not be able to so easily find things like this that the market has missed. And yet here it is - a profound and obvious mispricing, staring me right in the face.

Telecom Italia savings shares are just one of 15 opportunities I have uncovered this month souring global markets day and night for bargains.*** The pundits out there that would have you believe that markets are efficient or universally overpriced and devoid of bargain opportunities, are flat wrong.


*It is important, however, from the point of view of the investment merits of the savings shares. If the ordinary shares were materially overpriced, then that might serve to neutralise any benefits from the superior economic characteristics of the savings shares, as discussed. However, I do not believe that to be the case in the present instance, and believe the ordinary shares to themselves be moderately undervalued. 

**5% of 55c; the amount appropriated to the legal reserve is negligibly small, and the savings shares have, in fact, been paying 2.75c per year (excluding ordinary dividend entitlements) for many years.

***How did I find it? Well, telecommunications is out of favour at the moment, and Italy is also reasonably cheap. Looking at an Italian telecommunications company is therefore an eminently sensible place to look for a bargain. And no, I didn't read the entire 500pg annual report cover to cover - I jumped straight to the relevant sections. The notes on the company's equity capital structure, including the presence of any dilutive securities and differing share classes, is an important note you should always look at. 

Postscript: An investor friend of mine with a sharper eye than I picked up that the excess dividend entitlements accruing to the savings shares is 2% of 55c in excess of the ordinary dividend (or 1.1c), not 5% of 55c (2.75c) - see bullet two in the screenshot above (the wording is a bit contorted).

What this means in practice is that the savings shares are entitled to a dividend floor of 2.75c, but if the ordinary shares pay a dividend, the savings shares will only be entitled to a maximum of a 1.10c premium above the ordinary share dividend. For instance, of no ordinary dividend is paid, the savings shares pay 2.75c. If 1.00c is paid on the ordinaries, the savings shares still receive 2.75c. If 2.00c is paid on the ordinaries, 3.10c is paid to the savings shares. And if 3.00c is paid on the ordinaries, the savings shares are entitled to 4.10c, and so forth. 

These excess dividend entitlements should be worth closer to 30c per TIT savings share, in my current view, as compared to the 60c or more I originally referenced in the article. This is derived from capitalising the 1.10c perpetual excess dividend at 5-6%, which yields about 20c, coupled with a rough capitalisation of the likely excess dividends accruing above the 1.10c threshold (currently 1.65c per year) for a period of some years, given that the ordinary shares are unlikely to pay much in the way of dividends for some time, given the company's need to deleverage. Interestingly, the less dividends the ordinary shares pay, the higher the premium the savings shares justify (as the savings shares fully benefit from a reinvestment of retained earnings).

This additional nuance does not alter the original conclusions from the article: that the savings shares are materially mispriced relative to the ordinary shares. Even allowing for some discount for reduced liquidity and a lack of voting rights, they look to be worth roughly 90c to me - well above the current market price of 61c. This also rests on the assumption the ordinaries are merely fairly priced at 72c. As noted in the article, I also believe the ordinary shares to be somewhat undervalued.  

*The above analysis is furnished for informational/entertainment purposes only, and is not to be construed as investment advice. The author provides no warranty whatsoever as to the accuracy of the contents of the post, and reserves all rights to trade in any securities mentioned in any article at any time.