My Blog

November 28, 2007

WED 28 NOV 2007

The RMB declined today By Michael Pettis

After many days of strength, as of noon today the RMB declined by 0.15%, to 7.3949 to the dollar.  Is this simply a random event or are the Chinese financial authorities warning European officials to stop pushing on the RMB front? 

 

If it is the latter, it is likely to be a wasted warning.  Domestic pressures in Europe are going to continue to grow, and I expect the European trade deficit with China to rise even further over the next few months.  If we have more dollar weakness and a slowdown in the US economy, the European economy is going to have a real tough time adjusting, and already high unemployment is likely to rise.  Anti-Chinese feelings are already high in Southern Europe, and are likely to spread, and even in Germany there is a bit of a chill in the relationship thanks to Chinese anger at Merkel’s meeting with the Dalai Lama.

 

At this rate it is hard to see how China will be able to avoid a nasty dispute with Europe over trade – a French engineer I spoke with last night brushed off the nuclear and Airbus deals that China recently signed with France, saying that Sarkozy should not bow to such transparent ruses.  I wonder how many people in Europe see things in the same way.  At any rate there is almost no one left, even in China, who does not think the RMB is not seriously undervalued.  Fortunately there also seems to be a rising consensus in China that the RMB is not just a problem for China’s trading partners; it is an even worse problem for China.

 

Even though it makes sense for domestic reasons to adjust the currency more quickly, without serious foreign pressure I think Chinese authorities are still going to refuse, largely because of concerns about the possible adverse impact on domestic unemployment.  I think they are making a huge mistake, however.  China’s out-of-control monetary policy is already very likely to lead to domestic grief, and if we keep this pace of reserve accumulation up for another year, I think the chances of an ugly adjustment become extremely high.

 

Ironically, if US and European pressure is successful in forcing a currency move (trade sanctions would be disastrous for China’s export sector), it may still be too late to avoid a crisis anyway, only in this case the two countries would almost certainly be blamed for forcing the ensuing crisis onto China – in the same way the claim is often mistakenly made that the US “forced” Japan into the 1980s bubble.  In both cases domestic monetary policy was the culprit, but foreign pressure will be blamed.

 

WED 28 NOV 2007

Why IPO prices should surge, and what happens when they don’t By Michael Pettis

China’s largest heavy truck maker, Sinotruk, raised $1.2 billion in an IPO today in the Hong Kong market.  After its launch, however, the share price immediately fell by 16%.  This was the second China-related IPO in recent weeks whose price performance was so negative – last week Sinotrans Shipping raised almost $1.5 billion but saw its share price drop 13% after launch.

 

With these two exceptions, in the past months most China-related IPOs have seen their prices surge dramatically – in some cases by well over 100% – in the first day or two after launch.  One of the questions commonly asked in my finance class at Peking University is how do we justify the huge price run-ups we have seen in recent mainland IPOs.  Doesn’t that mean that the issuer got a bad deal by leaving too much cash on the table?  This is the same question that was often asked during the US internet bubble, when IPOs routinely shot up in value on the first trading day.

 

Aside from answers which imply fraud and insider activity, the only satisfactory answer I can give is based on the structure of the demand for and allocation of shares in any primary offering.  Basically, when an investor puts in a bid for a part of an IPO, his bid is firm (and in China 100% backed by cash), but there is no guaranty that he will receive the amount he bid.  If the deal is very “successful”, that is if it is vastly oversubscribed, the investor may get allocated as little as 1-2% of the amount he bid.

 

Basically that means that the buyer has given a put option to the arranger of the deal.  If I put in a bid for 100 shares, for example, I have given the arranger the right to sell me any amount from zero shares to 100 shares.  Of course the more successful the deal, and consequently the more allocation I want, the less I get, and vice versa.  If the deal turns out to have been overpriced, I am likely to get all, or nearly all, I asked for, much to my chagrin.

 

If investors are rational, this option should not be given for free.  In other words for me to justify giving the arranger the option I am granting him, I need to get paid, and the expectation of a run-up in share price after the IPO is what I am effectively being paid.  Imagine that there were no expected run-up in price.  In that case it would make no sense for me at all to participate in the primary offering – I should just wait until the deal is launched and then buy it in the secondary market.  This way I don’t take the risk of getting too little of an undervalued deal and too much of an overvalued one. 

 

So if we think of the bidder as granting an option, post-IPO pricing surges are not unreasonable.  Of course the more oversubscribed the deal, the greater the required pricing surge.  After all, I wrote an option on 100% of the shares I bid, but since I can only get a profit on the shares I am actually allocated, and I expect to get allocated very little unless the deal suddenly turns into a dog, the lower the expected allocation, the greater the expected pricing surge must be.

 

The initial post-IPO pricing surge, in short, is what investors are getting paid for providing an option to the arranger, and it is the arranging banks that enforce this payment.  By the way if this is true, we would expect post-IPO surges to be greater for more volatile assets (option premiums are higher on more volatile assets), and also greater the more oversubscribed the deal is expected to be (the premium on the full bid must be paid as a percentage a very small allocation).  Both predictions seem to conform well to the empirical evidence.

 

The recent slew of surging IPOs may make us forget that the risk provided by bidders in an IPO can be very substantial.  Imagine that I bid in a very hot deal but get allocated only 5% of my bid.  If the deal subsequently rises by 80%, this may seem like a huge profit, but it is actually a much lower return on my total capital at risk.  I have effectively received a profit equal to 4% of the nominal value of the options I granted, or 4% of the capital I put at risk. 

 

If the deal had turned out to be a dog, I would have been likely to get allocated a large part or even the full amount of my bid.  In the case of Sinotruk, I would have probably have been allocated all or most of my bid.  Assume that I got allocated50% of my bid (I have no idea of what the actual average allocation was), the 16% loss would mean that the options I granted immediately lost 8% in value.

 

The two recent unhappy IPOs indicate how risky the IPO market can be for investors.  In China, IPO investors typically bid for far more than they want because they expect to get allocated a very small portion of their bid.  In fact there is anecdotal evidence (and corroborating evidence in the performance of short-term interest rates) that many IPO buyers borrow as much money as they possibly can to finance their bids, and expect to sell quickly to repay their loans.  Should any IPO suddenly turn bad, however, these bidders are likely to be left with an excessively large investment in a plunging asset, and they will need to raise money quickly to pay off the debts incurred to buy the IPOs.

 

With the two recent badly-performing IPOs in the Hong Kong markets, I think the risk of poor performance in Shanghai or Shenzhen has risen substantially.  This may translate into unacceptably large losses for some unfortunate investors.

 

WED 28 NOV 2007

Quarterly loan quotas By Michael Pettis

According to a report I just received from Credit Suisse the PBoC has apparently begun asking commercial banks to submit quarterly loan plans for next year in order, ostensibly, to smooth out loan activity.  In the past, loan quotas were made on an annual basis.  Banks would then rush to make loans early in the year because they wanted to maximize interest income, but the result would be that roughly 40% of the annual quota (and typically more) was lent in the first quarter, and nearly all the rest by the third quarter. The banks then asked for quota increases and the authorities, faced with the prospect of a total loan freeze, would grant the new quotas.

 

Not surprisingly, credit growth always exceeded the PBoC’s already high annual targets.

 

Will this new tactic work?  I am skeptical, but I guess it depends on how you define success.  It may reduce excessive loan growth, but will it reduce the problems associated with excessive loan growth?  If we assume that the source of China’s explosive growth in industrial production, fixed asset investment, the trade surplus, and the speculative investment bubble is the unconstrained eagerness of Chinese bankers to lend money to their friends, then a measure to enforce a little discipline on those bankers can’t help but be a good idea.

 

But if the problem is uncontrolled money growth fueled by the country’s currency regime, than I am not sure how quarterly loan quotas are going to make much of a difference in the underlying problem.  It may discourage bank lending, but what about other effects?  Where does the money go?  If the commercial banks continue to gain deposits, they will have to put the money to work somewhere, and if they can’t lend the money they will be forced to purchase government bonds from the market.  This will keep interest rates low and of course will deposit the cash, which was otherwise to have been lent, into the hands of the previous owners of the government bonds, who will then be forced to something with the money.  Either way bank profitability should decline.

 

If the banks act in ways to discourage deposits, are those deposits likely to end up in informal banks who are, according to a lot of anecdotal evidence, a fairly large segment of the banking market and are largely unregulated?  Or will the money go into direct investment.

 

I am not sure how this plays out, but it seems to me if the problem is unconstrained money growth caused by massive capital and current account inflows, then the only way to address the problem is by reducing those inflows.