70. Book building is a non-underwritten form of placing. One or several book builders will pre-market a transaction and discover how much each investor will purchase of a bond at a given spread. For example an investor may buy £5m of a bond if the spread is 100 basis points but £7.5m if the spread is 110 basis points.
71. The book builders could be seen as finding the clearing price for the bonds for which they take no risk. However, a well run book building process can stimulate considerable enthusiasm and competitive pressure such that the clearing price improves over time. It has certain other advantages: book building is a marketing process where interest in the bonds is stimulated; a record can be kept of who is participating in the issue; a very detailed picture of the way the distribution process is going is built up over time and it can be adjusted accordingly - there should be no surprises; and fees are usually lower as the book builders' capital is not at risk.
72. Maybe the most important factor for the placing of a very large spread product issue is the management of the underlying Gilt market. As the book builders gradually discover the intentions of their purchasers, they can take steps to manage the flow of bonds into the Gilt market.
73. We believe the appropriate method to distribute the GGBs was book building although there was the risk that the clearing spread would have been wider. Many market participants were talking about wider spreads on the GGBs until very close to launch. This risk was perhaps also illustrated by the relative lack of success of the 2010 issue which was placed at a wider spread. We cannot, however, say that a better price would have been achieved via an underwritten placing; if anything we would be inclined to say that pricing would have been worse. This is partly because the underwriters would have sought to protect themselves against the risk of having to hold unplaced bonds. In addition, investors would have been fearful of poor performance of the stock if there was a chance that underwriters were having to hold large unsold positions.
74. It is difficult to analyse whether the final success of the long dated GGBs was as a result of the enthusiasm built up during the book building process or conditions in the Gilt market conducive to the placement of this type of issue. The more limited success of the 2010 issue illustrates that demand was less where the shortage of Gilts was not so severe.
75. Finally, we believe that the selection of book builders was important. Book builders were selected who are committed to the corporate bond market as well as the Gilt market in the UK. We believe the book builders would have been willing to buy any part of the GGBs still unplaced at the end of the book building process to ensure its success. It was a view in the market that this did in fact happen with the 2010 issue.