Sunday, January 17, 2010

Let me in! Let me in! I gotta get my welfare check! I mean my IPO allocation! Get outta the way!

In my previous post I wrote about the difficulties involved for Dubai to sell equity in its businesses in order to raise funds to pay down its’ debts. I have discussed already the psychology of Dubai and the reality that the level of indebtedness may exceed the amount of money that can be raised in equity. I would still argue that the strategic sale of equity is the best way for Dubai to manage the unwind of its’ debts. In this article I will address the most serious obstacle: the structure of the Emirati capital markets.

In order to explain the equity capital raising process it might be best to give an example of how it works in America. For that, I’d like to take the reader back to 1995 and the Netscape IPO. If you are too young to remember the early days of the internet Netscape was the first commercially available web browser. By August of 1995 virtually Netscape 80% of people using the internet were using Netscape to do so, so it had the best name recognition of any internet company. At the time people did not understand the commercial implications of the internet other than to know they were huge and so a lot of money poured into the field, initially from venture capitalists. The Netscape IPO was one of the first chances for the public at large to invest in the internet. Boy, did they ever.

The way that IPOs work in the US is that the company hires an investment banking advisor who does an initial valuation and gives a range. Then the company and their advisor go on a roadshow to pitch the company to potential investors and gauge their interest. Then a book building process is held which is kind of like a slow motion auction in which the investment bank collects orders from investors for the companies’ shares. When all the bids are in and the final price is decided the bankers go through and using their discretion to allocate the shares among the various investors. Usually they try to choose investors that are not likely to sell right away and occasionally “friends of the firm.”

What made the Netscape IPO so amazing was the investor demand for the shares. Netscape gave its product away and so it had little in the way of revenues as a result it was hard to value. Investors however gave the company the benefit of the doubt and bid the shares to the moon. Netscape offered to sell up to 5 million shares or about 10% of the company at a range of between $12-$14 dollars. During the roadshow it became clear that the demand was much greater so they upped the offer price to $28 dollars valuing the company at over a billion dollars. Even at this price there were over 100 million shares bid for meaning that the offer was oversubscribed 20 times with investors bidding for $2.8 billion. On the day the company went public the shares traded as high as $75 before closing at $58 and change. A home run for anyone lucky enough to get an allocation. Because of the massive oversubscription, many investors received no allocation and all those that did receive one got only a few thousand shares. MSAM the asset management wing of the lead bookrunner got only 6,000. Tom Brokaw got 28,000. Yep, people were mad about that. This massive initial valuation for an IPO touched off the internet boom in the United States that ultimately came to grief in 2000.

Just as Netscape kicked off my career the Dana Gas IPO kicked off my Middle Eastern career. Dana Gas, a start up gas company conducted an IPO on the Abu Dhabi Securities Market in October of 2005 in order to fund a massive expansion. The company solicited 300 “founders” (the equivalent of venture backers) who purchased 65% of the equity in the company. The remaining 35% was offered to the public. The IPO was ultimately 140 times oversubscribed with investors submitting 288 billion AED ($78 billion) in bids. The IPO price was 1AED and the stock opened at 5.48 AED a huge increase though like the internet boom Dana Gas has also come to grief with the shares closing at .98 AED today in Dubai.

So here we have two stories of two IPOs massively oversubscribed and massively successful on their first day of trading. Both were start ups, venture backed and were the tocsins of their eras. But what is important about them are not their similarities but their differences.

The first, and most obvious, difference is scale. The Netscape IPO doubled on its debut, Dana Gas more than quintupled. The oversubscriptions are very different, Netscape 20 times and Dana Gas 140 times. To say that one was 20 times and the other 140 times doesn’t go far enough to describe the difference. Not only was the Dana Gas oversubscription massive relative to the size of the offer, it was massive relative to the size of Dubai’s entire economy. In 2005 the total GDP of Dubai was 140 billion AED, so the amount bid for the Dana Gas IPO was greater than the value of all the goods and services produced in Dubai over the course of a year!!! How can that be!?! Imagine an IPO in the US generating $25 trillion in interest!! You might be thinking, surely you’re joking. No, I’m totally serious. Clearly this merits a closer examination.

The mystery behind this is that the IPO process in the UAE is very different from the process in the US. The most obvious way in which they differ is in the allocation process. As I have mentioned in America the IPO allocations are done in a discretionary process by the bankers. In the UAE the allocations are done formulaically and in two tranches. There is a retail tranche which is open to UAE or GCC nationals only. For the Dana retail tranche 700 million shares were set aside and divided evenly among all those who applied. For Dana that was about 400,000 people so each person got 1750 shares, thus each person made about 8000 AED on day one. Because of this allocation methodology the retail tranche can’t really be oversubscribed. What really happened was that the institutional tranche was 222 times oversubscribed, more than ten times Nestscape. That’s where the magic was.

While the retail tranche allocation is divided by the number of subscribers, the institutional tranche is divided by the value of each investors bid. If I bid for $1 million worth, and you bid for $10 million worth you get ten times as much as I do. This is novel and you could argue that it makes more sense than the American way of doing it where a handful of guys decide who gets what. In the UAE if you want more, you bid for more, you get more.

The banks add an interesting wrinkle to this: they’ll lend you the money to gross up your bid. HSBC was to lever some of their clients up to 5 times. So now, you put up one million, they lend you five million, now you’re in there bidding for $6 million and you’ll get a much bigger allocation. Many clients got significantly better leverage than five to one. I think the legal limit was 12 to one and some clients got even more than that. So why would the banks do this?

Well another difference in the UAE is that you have to actually put up the money at the time you subscribe for the IPO. As you don’t find out your allocation and get your refund for 30 days you have to pay interest on your loan for those 30 days. That’s how the bank makes money. It seems like a minimal risk to the bank, they’re lending you money to buy shares which have yet to be issued. If they lever you 10 to one and the issue is 20 to one oversubscribed they never even have to take a pledge of your shares because they get all their money back and you even get back some of your deposit. In the interim, your pledged money plus their loan to you go to the company which turns around and deposits it in the receiving bank until it’s refunded after the 30 days, so the money never actually leaves the bank.

Wait a minute, did you say the money goes to the company? Yes, the company gets to earn interest for 30 days on the amount of money bid for the IPO. So in the case of Dana Gas they got to earn interest on the GDP for Dubai for 30 days. I think the total proceeds to them were something like 1 billion AED, almost as much as they raised in the IPO itself. I’m not making this up. You’re probably thinking to yourself that something is missing here. What is the weak link in the whole argument? Well, there’s a big one.

This entire system is predicated two assumptions: 1.) there will be a massive increase in the shares on day one, and thus 2.) the issue will be massively oversubscribed. Imagine for a moment that the banks levered the subscribers 5 to 1 but the issue was only covered, not oversubscribed. Now the banks have essentially made margin loans to all their clients on a 5 to one basis at the IPO price. Unless the shares trade up a lot the bank is going to have to issue margin calls the instant that trading begins. If the shares decline then the bank is in very serious trouble indeed. Despite these risks banks and investors were willing to risk the entire GDP of the country on the fact that the Dana Gas IPO would be significantly underpriced and thus significantly oversubscribed. With a risk that big you would have to assume that it was a sure thing. So is it a sure thing that the IPOs in the UAE are systematically underpriced? Yes it is.

As I mentioned earlier in this article in the US the IPO price is determined by a process called book building where investors submit bids at varying levels beginning with an indication given by the bankers. This is done to try to arrive at a level which does not leave too much on the table for the sellers and at the same time leaves upside for the buyers. Who decides what the IPO prices should be in the Emirates? The government.

What happens is that the issuer and the advisors submit a request to ESCA, the Emirati Securities and Commodities Authority and ESCA has to approve the valuation. The ESCA valuations are always on the conservative side. There are several reasons for this. One is that ESCA, like the SEC has a mandate for investor protection. If you told the SEC to decide the value of all IPOs in the US it would probably make them cheap as well. Two, a failed IPO would look bad and the UAE will spend billions to not look bad. Most importantly is a philosophical reason and that is that the authorities look at the IPO process as a form of wealth redistribution. I have had government officials throughout the Gulf tell me this in person. The theory is that if you are running a successful commercial enterprise in the Gulf you probably had significant help from the state (not an unreasonable theory.) Therefore might it not be reasonable to expect the owners to take a hit on their IPO and share their good fortune with the regular folk by low balling the IPO price? ESCA thinks so and that's what matters.

So what are the implications for this? The most important one is that if you have a valuable business you would have to have a hole in your head to sell your business this way. Sure you might recoup some of the ESCA discount through the float on the oversubscriptions but the amount of that will be correlated to the size of the discount, the ease of subscription and the level of interest rates not the value of your business. As a result most of the businesses that have done IPOs were start-ups like Dana, government companies like the DFM or companies owned and managed by people who were subject to political pressure from the powers that be in the Emirates like Arabtec. It’s hard to see this by looking the the list of shares on the DFM, you have to look instead at the fact that the most successful businesses in Dubai would never ever consider doing an IPO.

To bring this back to my original point about the equitization of the debts of Dubai something it desperately needs to do. Local IPOs pose significant risks for Dubai. ESCA is a federal regulator in the UAE and therefore is controlled by Abu Dhabi. For Dubai to begin listing its crown jewels it would be subject to Abu Dhabi for valuations which would almost certainly be lower than it would get if a true auction were held. Gosh if only there were someplace that Dubai could sell equity where it would not be subject to the ESCA approval process and instead could do a book built IPO, wouldn’t that be a great idea. Well there is such a place it’s called the DIFC and it will be the subject of my next post.


Anonymous said...

the IPO process of Dubai Gov owned/operated entities could have 1 good side effect: at least some transparency and publication of their 'true' value , or at least what the owners and/or authorities think their true value is.
( I assume here that a proper auditor will examine the books and the financials will become public as part of the IPO process ).

Am I correct in this?

Rupert Neil Bumfrey said...

When were any of NasdaqDubai's primary equity listings based on "book building"?

With the upcoming merger of DFM and NasdaqDubai, which is perceived as a DFM take-over, will ESCA's role be as it is now with DFM?

Please continue the excellent output, as on my aggregated blog your posts get most hits, congratulations!

Anonymous said...


Definitely true...

lets see DP World ipoed at US$ 1.33, post the IPO Investors have seen spectacular returns (on an average) of -60%.

Surely the good govt of Dubai will price IPOs to generate such nice returns and surely institutional investors will folk towards such IPOs, Where else can they lose so much money so quickly

Ken said...

Thanks Rupert, KHI, DPW, DAMAS, and DEPA were all book built IPOs. As anonymous points out (and steals my punchline for my DIFX post) the Dubai government didn't leave much on the table for the DPW investors and killed the golden goose with a rich valuation. And of course the Abdullah brothers stole two thirds of the Damas IPO proceeds with no reprecussions so I think it may be too late for them.