Bad research KPN edition (and why my readers are a lot smarter than professional analysts)
This is “research” directly from Bloomberg:
KPN NA: 2.74 TARGET: 2.20-2.40
April 25, 2013
We recommend shorting KPN on the share issue announcement. The equilibrium price is Eur 2.04/share and the stock price should trend toward this level in the near future.
Equilibrium Price = (price pre-cap raising announcement x # shares + price cap raising x # shares) / total # shares.
The shares were trading at around 4.0 pre-announcement and there were 1.431 bn shares outs.
The company is issuing 2.84bn shares at 1.06. This works out 2.04/share. Let’s imagine that thanks to the cap raising the
fundamentals are slightly better now than they were before, and we set our target at 2.20-2.40, or 10-20% lower than the current price. The stock should be sold/shorted.
Let’s look at their “formula”:
Equilibrium Price = (price pre-cap raising announcement x # shares + price cap raising x # shares) / total # shares
“Wow, they are professionals and have a formula, this must be right” was my first thought and I briefly thought that I have made a mistake. Although I had a nagging feeling that the formula (or the application of it) was not right.
Enter readers JM and Martin in the comments of the last KPN post:
26. April 2013 um 09:57 | #8 Antwort | Zitat | Bearbeiten
ok…I join the game. When the capital increase was announced the share price was Euro 3,1…so 3,1+1,06+1,06/3 = 1,74. If the dust settles this share price is my personal target…more I do not expect .
26. April 2013 um 10:35 | #9 Antwort | Zitat | Bearbeiten
The day BEFORE the capital increase was announced, the stock price was 4.10 EUR.
26. April 2013 um 10:55 | #10 Zitat | Bearbeiten
price today 1,56
As there were cost for KPN real economic discount is somewhat lower.
Martin and JM (by the way, thank you for your many helpful comments) are using the very same formula, however with one big difference: They compare the result with the stock price AFTER the rights have been split of.
So the question is: Who is right ? The “professional” reasearch on Bloomberg or the comments on my amateur blog ?
Well, if you are a KPN shareholder before and KPN would sell the new shares directly to new shareholders at 1.03 EUR per share, then of course the old shareholders would be massively diluted.
In reality however, no one is able to by the shares directly at 1.03. You can only access those cheap new shares by buying the subscription rights which have been giving to the old shareholders as compensation for the dilution.
So the big mistake made by the “professional” analyst was that he somehow forgot to account for the subscription right. as the subscription right was worth 1 EUR pr share, suddenly an overvalued share according to the “professional” is an undervalued share.
How can a “professional” be so stupid ?
My guess is that the “analyst” mixed u a rights issue with an issue of shares without rights. Companies usually can issue a certain amount of shares (usually max. 10%) directly to new shareholders without any rights to old shareholders.
In those cases, the announcement is usually made shortly before the trading day starts and then the formula above is applied to the price the day before and the price for the new shares.
Nevertheless it is really embarrassing for any institution that such a mistake slipped through and this report gets actually published on Bloomberg
Applying the formula to a rights issue
The “formula” rests on 3 critical assumptions:
1) The price of the shares on the day before the announcement is exactly the right (and efficient) price. There was no information out there before about the capital increase
2) Nothing happened in between
3) the additional capital will not create any additional value
I think all three assumptions are quite difficult to hold for the KPN rights issue. Talk about the issue started already in September last year, so at the time of the announcement (Feb. 5th), the stock price reflected already a part of that.
Also, the time period between announcement and pricing of the issue is quite long with 6-7 weeks.
It is amazing, how bad some of the “professional” research actually is. In that case the analyst used a formula without accounting for the subscription rights. However this also shows that in those situations, the price discovery might be not fully efficient.
I wouldn’t put too much faith into the “formula”, as the application towards rights issues really stretches the implied assumptions.