|
|
Thursday, 25th April 2024 |
| Trends in IPO methods - how to get it right |
Back |
| The popularity of book-building for initial public offerings (IPOs) has grown in the last decade despite the greater costs involved. UCC academics Edel Barnes, Derek Beatty and Gael Hardie-Brown examine how it can be both successful and effective. |
Globally, the late 1990s were charcterised by very active markets in IPOs, the mechanism whereby organisations obtain a first listing of their shares. This period of very hot IPO activity was driven both by an increase in the number of technology and internet-based entities that sought funding and the lofty valuations afforded to these firms.
A number of methods are available to firms to effect an IPO. The particular method chosen depends both on the market environment into which the firm will sell its shares and characteristics of the firm itself. The key objective however is to successfully market all shares on offer and to optimise the proceeds of the sale, but therein lies a tradeoff.
Typically very little is known about start-up firms and from an investor’s perspective information acquisition is costly. If there is substantial uncertainty regarding a firm’s activities and prospects and resolving this uncertainty is associated with significant costs, investors will either not subscribe to a new issue so that an offer fails. Alternatively they will subscribe only if offered new shares at a discount which reduces the net proceeds to the issuing firm.
The offer price is thus key to ensuring a successful issue but if set too low, it results in significant under-pricing whereby investors earn high initial returns but where the opportunity cost to the issuing firm of forgone share capital can be considerable. The US, the hot issues market of 1999 and 2000, was associated with initial day returns of 65 per cent on average, and although returns were not so pronounced in other world markets, initial returns to the order of 20-30 per cent were not uncommon.
The various methods of bringing new issues to market differ largely in terms of their risk of under subscription and the costs they impose on investors in acquiring firm-specific information. Book-building, fixed offer price and auctions are the most commonly used IPO mechanisms, and while some countries are associated with use of just one mechanism, typically firms have a choice. Generally, public information is not completely incorporated into final offer price for IPOs and firm risk is positively related to the extent of under-pricing and/or initial returns to investors.
1. In book building, firms hire an underwriter to certify the new issue as regards firm quality and fair pricing. The fundamental assumption underlying use of this mechanism is that the underwriting firm has the best understanding of market conditions and access to potential investors. The underwriter will research the issuing firm, the firm and underwriter will agree a price range for the IPO and the underwriter will subsequently engage in a ‘road show’ to elicit non-binding indications of interest in the new issue from investors.
Once this period of book building is over, the underwriter and firm will agree a final offer price and the underwriter has complete discretion in the allocation of shares once this price is set. Notwithstanding the substantial fees charged by underwriters for this service, the mechanism is utilised by the vast majority of issuing firms worldwide and is used to the virtual exclusion of other mechanisms in the US. This is largely because it allows issuing firms to control spending on information acquisition and minimises the risk to the issuing firm of offer failure, which risk is borne by the underwriter.
The reduced risk of offer failure comes at a price; usually IPOs conducted by the book-building mechanism are associated with significant underwriting or direct costs. Incentives for underwriting investment houses to favour privileged clients in ‘hot IPOs’ and to price new issues attractively to guarantee high initial returns adds to the cost of utilising the book-building mechanism.
2. Under a fixed-offer price regime, the number of IPO shares and price at which these will be issued to the public is set and advertised approximately a week before the actual IPO date, this price being the result of negotiations between an issuing firm and its underwriter.
Potential investors submit their orders for new shares at this fixed price and shares are allocated on a pro-rata basis. Again underwriters incur costs of acquiring information about the issuing firm, which cost is levied on the issuing firm in the form of underwriter fees/compensation. Because offer price is set before information regarding investor demand is known, the underwriter bears the risk of under-subscription and must buy any unsold shares, so typically offer price will stand at a substantial discount to true value to ensure full subscription.
In consequence, fixed-offer prices IPOs are associated with greater under-pricing than other mechanisms on average, which imposes an indirect cost on the IPO firm. Because efficient price discovery requires some adjustment of offer price to demand, this fixed offer price mechanism is commonly regarded as one of the less efficient pricing mechanisms and is not extensively used in consequence.
3. IPO firms can bring their shares to market via a Dutch auction approach (sometimes referred to as a uniform price auction) where investors are invited to submit bids indicating both the number of shares required and the price they are willing to pay.
IPO shares are then sold to the highest bidders, with a uniform price set at the level of the bid of the lowest winning bidder. This standard uniform price equates supply of shares with demand by investors. Underwriters may engage in information acquisition before offer price is set but unlike the book-building mechanism issuing firms do not control spending on company research and investor demand determines expected issue proceeds.
Auctions also carry a much higher risk of under subscription than alternative approaches and the offer price that clears the market is generally well below fair value, particularly for companies and in industries that are not well established or understood. A further risk is that it may be possible for bidders to tacitly collude by placing demand functions such that market-clearing price is very low, if aggressive bidding to gain market share would push prices too high to yield attractive initial returns to investors.
4. In France, a variation on the Dutch auction approach, ‘offre ? prix minimal’, is used. Here the underwriter and issuing firm set a minimal acceptable offer price approximately one week before IPO date. On the date prior to issue, investors make price and quantity bids, which are collected and used to assess investor demand by the French market authority, the protector of investor interests. Underwriter and firm negotiate with this authority on offer and maximum prices based on this demand. The maximum price is chosen to eliminate unrealistically high bids. Collection of bids encourages investors to reveal their assessment of firm value, and shares are subsequently allocated on a pro-rata basis to bidders at a uniform offer price that lies between the acceptable minimum and maximum prices, which clears the market.
A logical question regards the type of selling and underwriting procedure that might be preferred for controlling the amount and volatility of under-pricing given that it imposes such a significant cost on firms coming initially to market by IPO.
A study by Loughran, Ritter and Rydqvist (1994) of the international evidence on IPO under-pricing reports an average level of under-pricing across 25 countries of 31 per cent over a variety of periods studied. Malaysia recorded the greatest under-pricing due to binding governmental constraints on the setting of offer price and France reported the lowest average initial returns of 4.2 percent. High initial returns were generally associated with regulated markets like Malaysia and those characterised by high marginal tax rates on income such as Sweden. Under-pricing was typically greater when offer price was set before information on investor demand was known as is the case for the fixed-offer price mechanism, and lower for auction-like mechanisms where the market clearing approach equates demand and supply. Book building was associated with intermediate under-pricing but imposed substantial indirect costs on issuing firms. Nevertheless, in excess of 90 per cent of IPOs adopted this approach.
The French Second March? is somewhat unique in that three basic and substantially different issuing mechanisms operate in juxtaposition: auctions, book-building and fixed offer price issues. Thus it merits a closer examination with a view to assessing the relative efficiency of the various approaches. Derrien and Womack (1999) report an overall average level of under-pricing of 13 per cent across all mechanisms for the French stock market as a whole, with average initial returns of 17 per cent, 9 per cent, 6.5 per cent, for book-building, fixed offer price and pure auctions respectively.
This suggests that pure auctions may be optimal for issuing firms concerned with minimising under-pricing. The French auction mechanism is market driven rather than underwriter driven and is associated with fewer frictions and reflects investor valuations more completely in the offer price relative to other approaches. Furthermore it was found to adjust most completely for recent market conditions in the pricing of IPOs, given that periods of high (low) IPO activity have been found to be associated with less (more) under-pricing.
It has been argued, and there is anecdotal evidence to the effect that the book-building approach to bringing new issues to market is inefficient largely due to the market power of underwriters and privileged investors who exercise significant control over the setting of offer price. Underwriters may allocate shares in a ‘hot’ new issue disproportionately to favoured investors in exchange for commissions and/or set offer price at a steep discount to ensure attractive initial returns to such investors.
Evidence on initial returns suggests that the auction-like approach, offre ? prix minimal, utilised in France may be most efficient in terms of minimising under-pricing. In consequence we might expect this approach to dominate in a market where a number of diverse issuing mechanisms are possible. Is there any evidence that (a) the approach actually dominates new issue activity and (b) that ‘French auctions’ represent the most efficient issue technology?
A recent study by UCC academics notes that while French auctions are indeed efficient in that they are associated with few of the conflicts of interest typically associated with book building, nevertheless book-building has been gaining in popularity in the French Second March?. The study examines 55 new issues, which were brought to the second march? during the period 1999-2001 inclusive.
Interestingly just four issues (7 per cent of all) used the pure auction approach, while 47 per cent or 85 per cent utilised either book building or a hybrid approach of book-building/auction or book-building/fixed offer. Pure book building accounted for just three new issues while 44 issues were hybrids. This pattern contrasts to earlier studies of the French market where in excess of 50 per cent of all new issues occurred by auction, and were associated with significantly lower under-pricing than other mechanisms.
Despite its pricing efficiency, the auction mechanism seems to be losing popularity over time, to the benefit of the more costly book-building-type approach. Nevertheless, over the period studied the French Second March? continued to be associated with lower average under-pricing than that documented for other markets where book building dominates.
A resolution of this seeming new issues paradox may lie in an ability of French IPO firms to benefit from the demand revealing attributes of book-building while avoiding the costly conflicts of interest associated with underwriter-investor client relationships. Were it possible to avoid such costly conflicts, it has been argued that book building could be just as efficient a pricing mechanism as the French auction-like approach. Certainly pure book building is an underwriter driven mechanism and too much pricing and allocation power in the hands of underwriters may result in substantial under-pricing.
However the UCC study makes the point that a greater proportion of French firms enjoy relationship banking than is the case in the US, UK and other markets, and French IPO firms tend to choose their main creditor bank as underwriter when going public. Thus underwriters know and understand the issuing firm prior to IPO, need to engage in considerably less information acquisition, are less likely to abuse the underwriting role with a client firm and will both require and choose less under-pricing to successfully market a new issue than in other global markets.
This reduction in under-pricing costs allows French firms to use hybrid issue mechanisms that incorporate both the beneficial attributes of book-building (research, investor road-shows, access to potential investors) and the demand-revealing auction or fixed offer mechanisms, in preference to the pure auction mechanism to attract foreign investors who are perceived to less fully understand the French auction-like approach.
This approach has resulted in average IPO under-pricing of just 6.3 per cent on the French second march? while the French nouveau march?, which utilises only pure book-building has been characterised by under-pricing to the order of approximately 17.5 per cent on average. The inescapable implication is that the French Second March? may be applying the book-building IPO method optimally, avoiding the conflict-related costs associated with other jurisdictions and continuing to enjoy significantly lower new issue mispricing. |
Edel Barnes and Gael Hardie-Brown lecture in the Department of Accounting and Finance, University College Cork while Derek Beatty, a former postgraduate student at UCC, now works in financial consultancy in Guernsey.
|
| Article appeared in the November 2002 issue.
|
|
|
