In early January 1996, Ms. Joanne Partridge, Director of Research at High Street Global Advisors (“High Street”), a Boston-based global investment management organization, was studying the price behavior of the shares of Royal Dutch Petroleum and Shell Transport and Trading. It seemed that Royal Dutch and Shell should trade in fixed proportions since they represented equivalent classes of shares of the same holding company. However, the ratio of share prices had been anything but constant. For example, Shell traded at a premium to Royal Dutch during 1990 and 1991, while Royal Dutch traded at a premium to Shell subsequent to 1991.
Presently, the premium of Royal Dutch over Shell was at an all-time high of almost 12%.
Joanne Partridge was trying to understand the opportunities presented by the Royal Dutch/Shell pricing discrepancy. Several of High Street’s U.S. domestic equity and global equity portfolios currently held significant positions in Royal Dutch. These positions could potentially be sold and replaced with equivalent-sized positions in Shell.
In addition, the firm had recently landed several new accounts, and would soon be investing the funds. It would have to decide whether these new accounts should own Royal Dutch or Shell. Finally, High Street managed a hedge fund, High Street Partners, which could attempt to arbitrage the price discrepancy by taking a long position in Shell and an offsetting short position in Royal Dutch.
High Street Global Advisors managed approximately $40 billion of tax-exempt assets for pension funds, foundations and endowments, and about $15 billion in mutual funds held by individual investors.
Most of these assets were in equity portfolios, whose investment mandates ranged from purely U.S. domestic to non-U.S. to fully global.
High Street viewed the world as consisting of one global economy. Accordingly, it emphasized appraising investment opportunities in a global context. At the core of the firm’s equity investment capability was a team of analysts who followed global industries such as chemicals, pharmaceuticals, autos, and oil, and who recommended their best stock selections within these industries to the
Professors Kenneth A. Froot and Andre F. Perold prepared this case. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Much of the data in the case is drawn from Kenneth A. Froot and Emil Dabora, “How are Stock Prices Affected by the Location of Trade?,” Harvard University, May 1996. Copyright © 1996 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School.
This document is authorized for use only in Speculation, Crisis & Behavioral Finance (Huberman) EMBA FA14 by Gur Huberman at Columbia Business School from July 2014 to January 2015. various equity portfolio managers. Partridge played a key role in giving direction to these analysts and in managing the flow of ideas between them and the portfolio managers. Portfolio management at High Street was generally governed by a value-investing philosophy according to which securities were purchased if their prices were attractive relative to underlying company fundamentals. In the case of Royal Dutch, the oil analyst was recommending the company on the basis of its lower price-to-book and price-earnings ratios than the major U.S. oil firms and because the company was contemplating certain refinery shutdowns and other operating restructurings that would improve its competitiveness.
Royal Dutch Petroleum and Shell Transport and Trading were not independent companies. The two were linked to one another by corporate charter, which mandated that cash flows to the equity holders of each company should be distributed in a 60/40 ratio. (See Exhibit 1 for balance sheets and income statements of the combined Group companies.) The companies stated that, “the RoyalDutch/Shell Group of companies has grown out of a 1907 alliance between Royal Dutch and Shell Transport by which the two companies agreed to merge their interests on a 60/40 basis while remaining separate and distinct entities.”
The organizational structure of the Group companies is depicted in Exhibit 2. All subsidiary companies’ shares were held by the Group Holding Companies, which in turn were owned by the two parent companies, Royal Dutch Petroleum and Shell Transport & Trading, in the ratio of 60/40. Royal Dutch and Shell were independently incorporated in the Netherlands and England, respectively. The companies’ alliance meant that all inflows from and outflows to shareholders were split 60/40.1 Combining this 60/40 split with the number of shares of Royal Dutch and Shell outstanding, meant that one share of Royal Dutch was entitled to the same cash flows as 9.2744 shares of Shell.2
The Group had attempted to make information widely available concerning parent company linkages. In addition to being explained at the beginning of each Annual Report, the corporate connections were detailed in 20F submissions to the U.S. SEC. The linkages were also the subject of a dedicated analyst/investor guide. While the Group actively attempted to split the cashflows according to the 60/40 ratio, there were a number of factors that caused dividend payments to deviate away from that ratio. These issues are discussed in the Appendix below. Analysts at High Street believed these factors to be relatively minor.
Royal Dutch and Shell were listed on nine exchanges in Europe and the United States. Most of Royal Dutch’s trading activity took place in the United States and the Netherlands markets, whereas Shell’s trading occurred predominantly in the U.K. market. In New York, however, Shell shares did trade as American Depository Receipts (ADRs), with one ADR being equivalent to six shares of Shell Transport and Trading. Thus, 1.5457 (9.2744/6) Shell ADRs were equivalent to one share of Royal Dutch. Geographical ownership information for Royal Dutch and Shell are shown in Exhibit 3. Exhibit 4 contains information on the trading volume of Royal Dutch and Shell in New York, 1″Royal Dutch and Shell Transport shall share in the aggregate net assets and in the net aggregate dividends and interest received from Group companies in the proportion of 60/40.
It is further arranged that the burden of all taxes in the nature of or corresponding to an income tax leveeable in respect of such dividends and interest shall fall in the same proportion.” Royal Dutch 20-F, 1993, pp. 1-2. Specifically, the company distributed corporate tax shields (generated by Shell’s dividends under UK tax law) on a 60/40 basis to the shareholders of both companies (see the Appendix below). 2As of January 1996, there were 536,074,088 shares of Royal Dutch and 3,314,503,242 shares of Shell outstanding. London, and Amsterdam since 1991. Royal Dutch had long been included in the S&P 500 and the most popular Amsterdam stock index, the CBS Herbeleggings. Similarly, Shell had long been included in the major index of U.K. stocks, the Financial Times Allshare Index (FTSE).3 Although Royal Dutch was a foreign-owned corporation, it was considered a U.S. stock by many institutional investors by virtue of its inclusion in common U.S. stock indexes. Major institutional holdings of Royal Dutch and Shell are listed in Exhibits 5 and 6.
It appeared that arbitrage across markets disciplined the price of Royal Dutch, so that it was essentially equal around the world. That is, at a given time, it would cost an equivalent amount to buy a share of Royal Dutch in Amsterdam as it would in New York. The same was true of Shell prices in London and New York, although for Shell the geographic disparities were generally somewhat larger. (See Exhibit 7 for historical geographic price disparities of Royal Dutch and Shell and Exhibit 8 for current pricing differentials.) However, the price of Royal Dutch fluctuated considerably when compared with the price of Shell. For example, on January 3, 1996, shares of Royal Dutch and Shell closed in Europe at fl227.8004 (Amsterdam) and ?8.6300 (London), respectively. At prevailing exchange rates, these prices were close to those that prevailed on the same day at the close of the New York markets.5 (See Exhibit 9.)
However, in both Europe and the United States, Royal Dutch was considerably more expensive than Shell (see Exhibits 8 and 9). Partridge was curious about the nature of the pricing differential. She wondered whether the strong performance of the S&P 500 compared to international stocks in 1995 might explain part of the current premium on Royal Dutch shares.6 In particular, Partridge wondered whether Royal Dutch would appear more highly correlated with the United States and Netherlands markets than Shell, and, similarly, whether Shell would appear more highly correlated with the U.K. market than Royal Dutch. If so, then an increase in, say, U.S. stocks would, all else equal, result in an increase in the price of Royal Dutch relative to that of Shell.
To investigate this, Partridge had an analyst compare the betas of Royal Dutch and Shell. The analyst regressed the difference between the returns on Royal Dutch and Shell on both market index and currency returns. (The betas of the Royal Dutch / Shell return differential are reported in Exhibit 10.) For example, a beta of 0.2 against the S&P 500 would indicate that a 1% increase in U.S. stocks (holding other countries’ stock prices and currencies constant) would be associated with a 20 basis point increase in the price of Royal Dutch relative to that of Shell. Partridge also knew that dividend withholding taxes might alter investor perceptions of relative stock value.
This should not have been very important for private investors in the United Kingdom, Netherlands, and United States, all of whom faced symmetric withholding taxes on the dividends of Royal Dutch and Shell. However, pension funds sometimes faced tax asymmetries with respect to the two stocks. For example, U.K. pension funds were exempt from withholding taxes on Shell, but not on Royal Dutch, and conversely, Netherlands pension funds were exempt from withholding taxes on Royal Dutch, but not on Shell. Partridge wondered whether tax issues could explain the behavior of the relative price of Royal Dutch versus Shell. Exhibit 11 shows the dividend and withholding taxes faced by different investor groups.
Analysis of the Investment Opportunities
Before suggesting any trades based on the price differential between Royal Dutch and Shell, Partridge wanted to better understand the costs that might be involved. To do so, she enlisted the help of High Street’s trading desk and also a prominent Wall Street firm through which High Street funneled much of its volume in international stocks, and which High Street also used for customized derivatives transactions.
Partridge began by thinking through the economics of selling Royal Dutch and purchasing Shell. This would be relevant for the portfolios in which Royal Dutch was presently being held, and which had a mandate for owning “foreign” as well as “domestic” shares. Some of High Street’s clients had given it strict “U.S.-only” mandates which permitted it to hold Royal Dutch but not Shell. On the New York Stock Exchange, both Royal Dutch and Shell ADRs were typically quoted at a 25? bid-offer spread in small-sized quantities (one thousand to five thousand shares). The spread usually would be wider for large sized trades. In addition, for trades in listed stocks, High Street paid its brokers a one-way commission of 5? per share. In Amsterdam, Royal Dutch was typically quoted at a spread of fl0.3 for small trades, and, in London, Shell was typically quoted at a spread of ?0.03 for small trades. In both Amsterdam and in London, High Street would pay one-way commissions on top of these spreads of 30 basis points.
The United Kingdom also imposed Stamp Tax, a 50 basis point transfer tax on purchases of U.K. stocks, including Shell.7 Trades in Royal Dutch in Amsterdam and in Shell in London would also require the conversion from guilders and pounds to dollars. These currencies tended to trade at bid-ask spreads of six basis points. If High Street’s hedge fund were to attempt to arbitrage the price discrepancy, one alternative would entail selling short shares of Royal Dutch and purchasing shares of Shell. In a short sale, the hedge fund would borrow shares of Royal Dutch, sell the shares, and later repurchase them and return them to the lender. The hedge fund would have to reimburse the lender for any dividends paid on Royal Dutch shares during this interim period. In addition, the hedge fund would have to pay a fee for borrowing the shares. This tended to run about 40 basis points per annum. This fee usually took the form of an interest rate give-up on the proceeds of the short sale.
The proceeds of a short sale would usually be held as cash collateral to protect the lender against borrower default. The cash would be invested in short-term instruments earning LIBOR or slightly less, and all but 40 basis points of this interest income would be rebated to the hedge fund. The hedge fund also would have to finance its long position in Shell. Presently, it was able to borrow at a rate of LIBOR + 75 basis points (on a fully collateralized basis). The hedge fund often took significantly leveraged positions, especially in situations where the risk was deemed minimal.8 There were other alternatives available to the hedge fund involving the use of derivatives.
One set of strategies involved the use of exchange-traded put and call options. There were fairly active options markets for Royal Dutch in the United States and Shell in the United Kingdom. These were short-term options, however, with maturities of six months or less. The options were usually slightly cheaper to trade than the underlying shares, although with rollovers they would become more expensive. A potentially attractive feature of options-related strategies was that they permitted the hedge fund to easily tailor its risks in an asymmetric fashion.7 No Stamp Tax was levied on purchases of Shell ADRs, however. 8 On its hedge fund, High Street received a management fee of 1% per annum on net assets plus 20% of profits earned in excess of LIBOR.
A second derivatives-related strategy involved the use of a privately-negotiated total return swap. The simplest total return swaps involved two counterparties agreeing to exchange the total return on one instrument for the total return on another, plus or minus a fee, where the total return on an instrument is its price appreciation or loss during the holding period, plus interest or dividend income paid on the instrument during the holding period.
Swaps could be customized in almost any way, for example, to allow the exchange of total returns on differing underlying amounts (“notional amounts”), or to include option-like features. Swaps also could be written for just about any maturity. The fee charged by the issuer of the swap typically depended on the ease of trading in the underlying instruments, the use of options and other features which might be embedded in the swap, and the credit-worthiness of the counterparty.