Pricing and Offer Methods in SIPs

As we will see in chapter 7, the past dozen years has witnessed an unprecedented surge in the number and value of corporate security offerings around the world.

table 6.4. Pricing, Share Allocation, and Control Allocation Patterns in Share Issue Privatizations

Initial SIPS

Seasoned Offers

Initial SIPS

Seasoned Offers

Measure

Mean

Median

Number

Mean

Median

Numbe

Pricing variables

Issue size (US$ millions)

$555.7

$104.0

417

$1,068.9

$311.0

172

Initial return

34.1%

12.4%

242

9.4%

3.3%

55

Percent of offer at fixed price

85.0%

100.0%

273

61.0%

100.0%

77

Cost of sales, percent of issue

4.4%

3.3%

178

2.5%

2.6%

61

Share allocation variables

Percent of offer allocated to

8.5%

7.0%

255

4.8%

2.6%

76

employees

Fraction of offers with some

91.0%

255

65.8%

76

allocation to employees

Percent of offer allocated to

28.4%

11.5%

348

35.9%

32.5%

142

foreigners

Percent of offers with some

57.1%

348

67.6%

142

allocation to foreigners

Control allocation variables

Percent of capital sold in of-

43.9%

35.0%

384

22.7%

18.1%

154

fer

Percent of offers where

11.5%

384

0

154

100% of capital sold

Percent of capital where

28.9%

384

8.4%

154

50% or more of capital

sold

Source: Data from Privatization International database, as reported in Tables 3, 4, and 5 of Jones, Megglnson, Nash, and Netter (1999).

Source: Data from Privatization International database, as reported in Tables 3, 4, and 5 of Jones, Megglnson, Nash, and Netter (1999).

For example, the total value of all security offerings was a little over $500 billion in 1990, but in 2003, the value was $5.3 trillion! Offerings by U.S.-based issuers always account for over two-thirds of this total, while European and Japanese issues account for most of the rest. Ljungqvist, Jenkinson, and Wilhelm (2003) document that U.S. investment banks — and especially U.S. banking procedures — have been steadily gaining market share since 1990, and the most visible aspect of this growth has been the spread of book building techniques. This means that an investment bank underwriting a share offering will carefully assess investor demand, at each of a rising schedule of per-share offering prices, by institutional investors for shares in a company going public. The bankers will then set the final offering price at a level slightly below that required to clear the market, then they will execute the offer by allocating shares to the institutional investors, after which trading begins. Book building, though expensive in terms of direct costs (such as underwriting fees), allows the issuing firm to sell shares at a sufficiently higher price that the net proceeds from the offering are higher than would otherwise be possible.

The alternative to book building is a fixed-price offering. These have traditionally been the dominant approach to pricing and marketing IPOs outside the United States. In this method, the issuer states the number of shares and the offer price several weeks in advance of taking orders from potential investors. In contrast, book building has been the method of choice in the United States. In this process, the investment banker builds a book as the issue is pre-sold to investors. This pre-selling involves the solicitation of interest but does not legally require investors to buy shares at the price they specify (though reputational concerns generally prompt investors to follow through and place firm orders). Book building has the advantage of aggregating investor demand prior to the offer and thus yields higher expected offer proceeds than the fixed-price method. However, Ben-veniste and Wilhelm (1997) show that fixed-price offers have an advantage over book building in that they are less likely to fail at the offer price. In fact, the fixed price is set lower than a book-building offer price so as to generate cascading demand that guarantees the issuers' proceeds. This makes fixed-price offers popular with risk-averse issuers, and it follows that these offers are appropriate for SIPs, since a failed offer could endanger the whole privatization program. In addition, cascading demand is useful for eliciting the interest of the politically important median-class voters. However, for less politically sensitive investors, such as institutional investors or foreigners, or for less politically sensitive seasoned offers, governments should be more willing to use book building or auctions to maximize issue proceeds.

The results presented in table 6.4 illustrate the popularity of fixed-price offers. For the 273 initial SIPs for which JMNN can determine the method of offer, 85 percent are at least partially fixed price and 79.9 percent are entirely fixed price. There are offers that are only partially fixed price, in that they use auction pricing for institutional investors. In seasoned offers, shares are sold exclusively at a fixed price 59.7 percent of the time, and 61.0 percent of the average seasoned issue is fixed price. Table 6.4 also presents mean and median values for the cost of selling the share offering as a percent of the issue amount. It is interesting to note that the average level of fees is only 3.9 percent across the full sample of SIPs, and the

4.4 percent average fee for initial offers is only slightly larger than the 2.5 percent value for seasoned issues.

These fees are much lower than the 7.0 percent and 7.9 percent spreads documented for recent American IPOs by Beatty and Welch (1996) and Lee, Lochhead, Ritter, and Xhao (LLRX; 1996), respectively. The fees on seasoned offerings are also much less than the 5.44 percent gross spreads documented by LLRX. Even when JMNN focus exclusively on the very largest ($500 million and up) initial offers in LLRX, they find that the 5.21 percent spread for the largest American IPOs is 18 percent larger than the spread for all initial SIP offers, even though the average size of SIPs is smaller than the top decile of U.S. offers. Evidence also suggests that the fees for larger SIPs are actually falling over time. The low costs of underwriting are consistent with governments deliberately un-derpricing to such an extent that the underwriters willingly accept low spreads because they perceive there to be very little risk of insufficient demand for shares. In addition, underwriters often view participation in a SIP as reputation enhancing.

Table 6.4 also reports the initial returns (i.e., the underpricing) calculated from the closing secondary market price on the first day of trading, less the offer price, as a percent of the offer price. The mean (median) level of returns is 34.1 percent (12.4 percent) for the initial SIPs and 9.4 percent (3.3 percent) for seasoned SIPs. JMNN find that underpricing has not decreased with time, and also document that the mean and median levels of underpricing are similar across industries with the exception of financials and utilities. We will discuss the theory and empirical evidence on underpricing in much greater detail in the next section. For now, we simply wish to document that underpricing of SIPs is pervasive, and assert that it is politically motivated. By deliberately underpricing SIPs, governments turn these share offerings into political instruments, since they can then allocate the underpriced shares (and implied capital gain) to politically favored groups of citizen/investors. As we now see, governments do in fact allocate shares based on a political calculus, and do so with great effectiveness.

Was this article helpful?

0 0

Post a comment