It is after midnight in the hedge fund oasis of Dubai and I am waiting to fly to Osaka. When I arrived here I got long of Dirhams at Citibank and then spent most of them in Abu Dhabi. I was going to exchange the balance back to USD until I came to my senses and bought some more - why does the UAE still have a dollar peg and let Fed decisions export inflation to a booming economy?
Coincidentally it seems the past week there has been a raging discussion on the "high" cost of capital that Citigroup (C) raised from the Abu Dhabi Investment Authority.
As best I can do from an airport cafe at 2.00am hacking around in Excel, I think this chart shows the payoff of the Citi FORWARD SALE of equity to ADIA.
A quick look at the Upper DECS equity units termsheet seems to reveal a fairly plain vanilla mandatory convertible. Yet media and blog commentary ranges from Citi forced into paying higher than junk to Citi got a good deal. This debate nicely illustrates how differences of opinion on the valuation of hybrid debt/equity financings and pricing disagreements on embedded options are all opportunities for alpha.
Financial engineering, model arbitrage and misunderstanding of what a particular security really represents allow skilled investors to make money out of the unskilled. The fact that many jumped on the "junk" 11% headline is a lesson in itself. That they missed the optionality is another. Subprime CDOs also paid a "high" yield but the buyers didn't realize all the credit default options they were implicitly writing much too cheaply.
The structure looks to me that Citi bought the right to sell stock to ADIA at 31.83 and sold a 17% out of the money call to ADIA so they could participate in any upside above 37.24. Since bank stock puts are massively bid by those seeking protection and implied volatility is high, ADIA picks up plenty of vol skew by writing a put and buying a call and so is obviously owed premium on such a collar. Therefore the SHOCKING 11% rate is explained as compensation for ADIA forgoing the 7% dividend yield available from Citi common stock, the high embedded option value that Citi purchased, the typical equity premium offered on a mandatory CB and the tax treatment.
The deal looks pretty fair to both sides. Citi probably minimized its cost of equity capital in a difficult situation while ADIA got what may prove to be cheap equity in a bank that is almost certainly considered by the US authorities to be too big to let fail. The 11% coupon convertible to a 4.9% stake were EXACTLY the same numbers offered in the convertible preferred bought by Prince al Waleed bin Talal the last time Citi got itself into trouble.
The credit cycle repeats but lessons aren't learned; SIVs and CDOs now, real estate loans and busted LBOs last time (sounds familiar), and "countries don't go bust" the time before that. Since billions and trillions get confusing, suppose an investor with a $90,000 portfolio decided to risk $750 on the stock of a major bank in distress and whose stock was far below its high. I don't think anyone would see that as particularly risky. ADIA has risked about 0.8% of its growing portfolio on the trade. Even if Citi declared bankruptcy the day after deal close and the stock went to zero, it would NOT be a disaster for the ADIA. They've written a put at a level they would presumably be happy to own the stock anyway and get paid a dividend (and the embedded option premium) until they take formal ownership of the stock in 3 years or so.
Panic is high, vol is high, vol skew is high, the current dividend yield "looks" high; why not bet a small proportion of your portfolio taking the other side?What other financing choices did Citi have? Why not a domestic investor like Berkshire Hathaway (BRK.A) who does have the cash instead of a sovereign wealth fund?
My guess is Warren Buffett did not want to tempt fate from a similar situation he had ages ago with Salomon, now part of...Citigroup. In September 1987, as a "white knight" he bought a Salomon convertible preferred with a similar "high" dividend (9% then) and "low" conversion price. Interestingly the deal was done with the stock at 31 and convertible at 38, almost the same situation as now! A few weeks later came the October crash of that year, obviously taking the stock way down. The price subsequently climbed the next few years ALMOST to the strike. But then the Treasury Bond scandal erupted hurting Salomon stock once again. Ultimately it was a reasonable trade for BRKA but I doubt he had an appetite this time.
Actually there probably was a "cheaper" financing source for Citi. Hedge funds might have been interested. After all Citadel just poured $2.5 billion into E-Trade (ETFC). And RAB Capital and SRM Global tried to catch that falling knife called Northern Rock (NHRKF.PK). With CBs, trading the optionality and mispricings in large bank mandatory convertibles has been quite lucrative over the years. When time ran out for the Japanese citybanks like Mizuho or Sumitomo-Mitsui in 2002/3, convertibles were issued for similar reasons and turned out to be both good investments and good trading vehicles for volatility monetization.
We may yet see a hedge fund targeted convertible emerge, though Citi will be glad to have gotten $7.5 billion done with a stable investor who won't short the common as a hedge. So much "pre-hedging", "pre-bookbuilding" and "wall-crossing" goes on that it would have forced the equity price down even more on an imminent large CB deal. Less negative effect on the stock price is a reason mandatory CBs are preferable for the issuer than ordinary CBs.With a bit of financial engineering and restructuring of cashflows the 11% coupon could probably have been as extreme as 0% or 20% by changing various preferential rights, the capital structure, optionality, convert strikes or other aspects of the deal.
What would the headlines have said then? "0% - Massive vote of confidence in Citigroup as it borrows at 400bp less than the US government?" or "20% - Citigroup forced to borrow far above subprime rates as bankruptcy looms?". The reality is that this was effectively a forward sale of equity, NOT debt finance. And the structure that got done was what suited both sides. Probably quite a few term sheets got proposed and items added or lined out before both sides agreed. Blog opinion ranges on the pricing from 11% junk to a somewhat more accurate Libor+150bp.
I think both Robert Rubin and Sheikh Ahmed Bin Zayed Al Nahayan knew what they were doing.While the experts say markets are becoming efficient and opportunities getting arbitraged out, it is reassuring to see disagreement on this, and in fact, every kind of financial security. The more complex, interconnected and innovative global investment products get, the more skill is required to understand and value them. That is why it is always going to be worth paying 2 and 20 to those able to identify mispricings in the markets and capitalize upon them. And why alpha really is portable; it transports itself from the many market participants that think they understand to the few that really do. Quality expertise costs more in any profession.
Anyway my flight is boarding. Must get to Japan since I am speaking on a conference panel about what Japanese investors want from a hedge fund. Not that it is a difficult question. It doesn't matter what passport someone holds or whether they are institutional, high net worth or low net worth; the investor requirements are basically the same. Reliable absolute returns EVERY year with minimal drawdowns and volatility, performance that MORE than compensates for the risk and fund managers who can tell at a glance if a security is ultimately debt or equity and have the models and experience to value it correctly.