【導(dǎo)讀】ChapterOpener. PART  III. p.381. p.382

　　

【正文】 o arbitrage.“Siamese Twin” Companies12 In 1907, Royal Dutch Petroleum and Shell Transport merged their operations into one firm. The two original panies, which continued to trade separately, agreed to split all profits from the joint pany on a 60/40 basis. Shareholders of Royal Dutch receive 60% of the cash flow, and those of Shell receive 40%. One would therefore expect that Royal Dutch should sell for exactly 60/40 = times the price of Shell. But this is not the case. Figure shows that the relative value of the two firms has departed considerably from this “parity” ratio for extended periods of time.Doesn39。t this mispricing give rise to an arbitrage opportunity? If Royal Dutch sells for more than times Shell, why not buy relatively underpriced Shell and short sell overpriced Royal? This seems like a reasonable strategy, but if you had followed it in February 1993 when Royal sold for about 10% more than its parity value, Figure shows that you would have lost a lot of money as the premium widened to about 17% before finally reversing after 1999. As in Example , this opportunity posed fundamental risk.Equity CarveOuts Several equity carveouts also have violated the Law of One To illustrate, consider the case of 3Com, which in 1999 decided to spin off its Palm division. It first sold 5% of its stake in Palm in an IPO, announcing that it would distribute the remaining 95% of its Palm shares to 3Com shareholders 6 months later in a spinoff. Each 3Com shareholder would receive shares of Palm in the spinoff.p. 389Figure Pricing of Royal Dutch relative to Shell (deviation from parity)Source: O. A. Lamont and R. H. Thaler, “Anomalies: The Law of One Price in Financial Markets,” Journal of Economic Perspectives 17 (Fall 2003), pp. 191–202.Once Palm shares began trading, but prior to the spinoff, the share price of 3Com should have been at least times that of Palm. After all, each share of 3Com entitled its owner to shares of Palm plus an ownership stake in a profitable pany. Instead, Palm shares at the IPO actually sold for more than the 3Com shares. The stub value of 3Com (., the value of each 3Com share net of the value of the claim to Palm represented by that share) could be puted as the price of 3Com minus times the price of Palm. This calculation, however, implies that 3Com39。s stub value was negative, despite the fact that it was a profitable pany with cash assets alone of about $10 per share.Again, an arbitrage strategy seems obvious. Why not buy 3Com and sell Palm? The limit to arbitrage in this case was the inability of investors to sell Palm short. Virtually all available shares in Palm were already borrowed and sold short, and the negative stub values persisted for more than 2 months.ClosedEnd Funds We noted in Chapter 4 that closedend funds often sell for substantial discounts or premiums from net asset value. This is “nearly” a violation of the Law of One Price, because one would expect the value of the fund to equal the value of the shares it holds. We say nearly because, in practice, there are a few wedges between the value of the closedend fund and its underlying assets. One is expenses. The fund incurs expenses that ultimately are paid for by investors, and these will reduce share price. On the other hand, if managers can invest fund assets to generate positive riskadjusted returns, share price might exceed net asset value.Lee, Shleifer, and Thaler14 argue that the patterns of discounts and premiums on closedend funds are driven by changes in investor sentiment. They note that discounts on various funds move together and are correlated with the return on small stocks, suggesting that all are affected by mon variation in sentiment. One might consider buying funds selling at a discount from net asset value and selling those trading at a premium, but discounts and premiums can widen, subjecting this strategy too to fundamental risk. Pontifff15 demonstrates that deviations of price from net asset value in closedend funds tend to be higher in funds that are more difficult to arbitrage, for example, those with more idiosyncratic volatility.p. 390Closedend fund discounts are a good example of socalled anomalies that also may have rational explanations. Ross demonstrates that they can be reconciled with rational investors even if expenses or fund abnormal returns are He shows that if a fund has a dividend yield of δ, an alpha (riskadjusted abnormal return) of α, and expense ratio of ?, then using the constantgrowth dividend discount model (see Chapter 18), the premium of the fund over its net asset value will be If the fund manager39。s performance more than pensates for expenses (., if α ?), the fund will sell at a premium to NAV。 otherwise it will sell at a discount. For example, suppose α = .015, the expense ratio is ? = .0125, and the dividend yield is δ = .02. Then the premium will be .14, or 14%. But if the market turns sour on the manager and revises its estimate of α downward to .005, that premium quickly turns into a discount of 27%.This analysis might explain why closedend funds often are issued to the public at a premium。 if investors do not expect a to exceed ?, they won39。t purchase shares in the fund. But the fact that most premiums eventually turn into discounts indicates how difficult it is for management to fulfill these CONCEPTCHECK3Fundamental risk may be limited by a “deadline” that forces a convergence between price and intrinsic value. What do you think would happen to a closedend fund39。s discount if the fund announced that it plans to liquidate in 6 months, at which time it will distribute NAV to its shareholders?Bubbles and Behavioral EconomicsIn Example above, we pointed out that the stock market runup of the late 1990s, and even more spectacularly, the runup of the technologyheavy NASDAQ market, seems in retrospect to have been an obvious bub