home air conditioners, refrigerators for trucks and stores, and other climate control products.
But some Ingersoll-Rand shareholders, who had expected the cash-rich company to pour some money into share repurchases, seemed disappointed with the acquisition announced Monday and sold Ingersoll-Rand stock, driving shares down sharply.
Merger arbitrage is attractive to many investors as a portfolio diversifier because of its long/short components. It is assumed that these positions immunize the portfolio against fluctuations in the overall stock market and leave only uncorrelated event risk to the investor, and therefore, the portfolio is market neutral. This argument is revisited in more detail in Chapter 3. Nevertheless, at this point, a short discussion of one of the pitfalls of long/short positions is necessary. A constant percentage spread can lead to dollar paper losses in an extreme bull market, if both the long and the short position increase, but the percentage spread remains constant. Table 2.2 illustrates the problem of a hypothetical increase of CGA and B2Gold when the percentage spread remains constant throughout the increase. The losses discussed here are temporary only and will eventually be recovered once the merger closes.
Table 2.2 Losses Suffered at a Constant Percentage Spread in a Rising Market
Table 2.2 starts in the first row with the actual spread of 2.20 percent at prices of C$2.86 and C$3.95 for CGA and B2Gold, respectively. It shows the profit and loss (P&L) relative to a position entered at the baseline of C$2.86 and C$3.95. If both CGA and B2Gold rise and the percentage spread remains constant, then the spread expressed in dollars must rise (2.2 percent of $5.72 is more than 2.2 percent of $2.86). The simulated price rise in Table 1.3 shows that a spread of $0.063 will widen to $0.126 per share if CGA were to double in value to $5.72 per share. Although B2Gold's stock appreciates by the same percentage as CGA, the difference in dollar terms increases. At $5.72 per share, the arbitrageur's portfolio would record a loss of $0.063 per CGA share (last column).
This scenario does not imply inefficiency in the market. If the hypothetical increase in spreads were to occur on the same day as the position was entered, the annualized return would be unchanged, because the percentage spread is the same whether CGA trades at $5.72 or at $2.86.
It is clear that these losses are only paper losses that are temporary. As long as the merger eventually closes, the arbitrageur will realize a gain of $0.063. Only those who panic and close their position early will suffer a real loss. The arbitrageur is short 0.74 shares of B2Gold for every long position of CGA, and the cash changed hands already when the trade was made. Therefore, the eventual profit is certain as long as the merger closes. In the meantime, however, the account will show a loss.
Whether or not an arbitrageur wants to hedge against paper losses is a matter of personal preference. Any hedging transactions will entail costs and will reduce the return of the arbitrage. Because the spread eventually will be recovered, it makes little sense to hedge against transitory marked-to-market losses.
Now what would happen if stock prices were to fall? It can be extrapolated from this discussion that in the case of a fall in stock prices, the dollar spread will tighten, and the arbitrageur will record a gain even though the percentage spread and the annualized return would remain unchanged.
Sometimes shareholders hold a number of target shares that does not get converted to a round number of buyer shares. For example, a holder of 110 shares of CGA would receive 81.4 shares of B2Gold. However, the fractional 0.4 shares cannot be traded or issued because corporations have whole shares only. (Note that mutual funds are different, even though they are also organized as corporations.) Therefore, companies will liquidate fractional shares and issue only full shares. The investor in our example would receive 81 shares of B2Gold and a cash payment for the value of the fractional 0.4 shares. The cash payment depends on the share price of B2Gold at the time of the closing of the merger.
In addition to earning the spread, a stock-for-stock merger has another source of income. When arbitrageurs short a stock, they receive the proceeds of the short sale. In the example from Table 2.1, the arbitrageur received C$2,923 from the short sale of B2Gold. These funds are on deposit at the brokerage firm that executed the short sale. Arbitrageurs can negotiate to receive interest on this deposit. This is easier said than done. In the author's experience, most retail brokerage firms do not pay interest on the proceeds of short sales. At the time of writing, one retail brokerage firm advertised that it had paid interest on balances of short proceeds in excess of $100,000. Institutional investors are better off. They are always offered interest on the proceeds. This is referred to as short rebate in industry parlance.
The example of the CGA/B2Gold merger can illustrate the effect of the short rebate on merger arbitrage returns. Assume that the short rebate is 1 percent. At the time of writing, in a period of historically low interest rates, this would be a high rate. In normal interest rate environments, short rebates are higher and match LIBOR rates. In fact, it is quite normal for rates for short rebates to be below interest rates. In fact, the spread between short rebates and margin rates charged customers who borrow to buy stock is an important source of revenue for brokerage firms. The interest earned on the $2,923 over the 140-day period until the closing of the merger would have been
2.7
This would increase the merger profit from $63 to $74.21 – an increase of almost 18 percent. For simplicity, simple interest is used in this calculation. Most brokers pay interest monthly, so monthly compounding should be used.
The annualized spread increases by the amount earned on the short rebate:
where
As discussed in Chapter 3, returns on merger arbitrage tend to be correlated with interest rates as a result of the impact that short rebates have on spreads.
The CGA/B2Gold merger was easy to analyze because neither stock pays any dividends. Stocks paying dividends can be tricky to handle when sold short, because the short seller must pay the dividend on the stock. The long position will generate a dividend; the short position will cost a dividend. A crude calculation to determine the net effect of dividends on the annualized spread is to subtract the dividend yield of the short position from the dividend yield of the long position, and add the result to the annualized return of the merger arbitrage. However, this method can give incorrect results, especially for mergers with a short horizon to closing. The method can be used as a first approximation, but arbitrageurs always must consider the actual dividend dates and dividend amounts.
The gross return in the presence of dividends is calculated for a long/short merger arbitrage in this way:
2.8
where
Mixed Cash/Stock Mergers
Many buyers want to limit dilution in the acquisition of a target company or have access only to an amount of cash insufficient to purchase the target entirely for cash. They structure the acquisition of a target for a dollar amount plus shares, or they offer target shareholders the option to choose between cash and stock, typically with
