Friday, May 28, 2010

Cash and Cross holdings

I am sorry about the long break between posts but I was on the road for much of the last two weeks and grading exams prior to that. I started my road trip with sessions in Slovenia and Croatia and continued on to India to make a presentation to the Tata Group, one of India's premier family groups, with a long history of operating in every aspect of Indian business. As part of the preparation, I did value a Slovenian pharmaceutical company (Krka), a Croatian tobacco company (Adris Grupa) and four Tata companies (Tata Chemicals, Tata Steel, Tata Motors and Tata Consulting Services). The presentations and the spreadsheets containing the valuations are online and can be accessed by going to:
http://pages.stern.nyu.edu/~adamodar/New_Home_Page/country.htm

Without belaboring the details, there were two key issues that came up when valuing Adris Grupa and the Tata Companies.

1. Cash holdings: Adris Grupa, as a tobacco company with significant operating cash flows, has accumulated a very large cash balance; it amounts to 20% or greater of the overall value of the firm. Adris is clearly not the only company that accumulates cash and it is not a phenomenon just restricted to emerging markets. Technology companies in the United States, such as Apple and Microsoft, have also been avid cash accumulators. While the conventional valuation practice with cash has been to add the cash balance to the value of operating assets, thus adopting the common sense rule that a dollar in cash has to be worth a dollar, there is substantial evidence that markets do not always treat cash as a neutral asset. In particular, markets seem to view companies that generate poor returns on their operating assets (less than the cost of capital) and accumulate cash with disfavor, while being much more sanguine about companies with good investment track records and substantial cash. Apple, for instance, is clearly not being penalized (and may be even be rewarded) for its large cash balance; after the most recent decade, investors trust the company to find good uses for the cash. In the Adris valuation, one of the concerns I raised was that the company's return on capital has lagged its cost of capital.It is therefore possible that the market may be discounting the cash holdings; a Croatian kuna in cash may be valued at less than a kuna.


2. Cross holdings: The Tata companies that I valued, with the exception of Tata Consulting Services, shared a common feature. A third to half the value that I estimated for each company came from  holdings in other Tata companies. In effect, investing in any Tata company is a joint investment in that company and a portfolio of 25-30 other Tata companies. While one reason for this cross holding structure is corporate control - it allows the family to preserve its control of the group companies - there are also more benign reasons, rooted in history. In the decades before the 1990s, Indian investors had little access to financial information from the company, let alone analyst reports or investment analysis. In that period, these investors had to essentially buy companies based on how much they trusted the promoters of the company, and a trusted family name became a proxy for research. In addition, when capital markets are undeveloped, having an internal family group capital market, where excess cash at some companies can be redirected to other companies that need the cash can be a competitive advantage.

The Indian equity markets today are different. While Indian companies have their own share of scandals and investment advice/ equity research can be tainted, the market is wider (thousands of publicly traded companies) and much deeper (more investors both from Indian and from outside). The cross holdings at family group companies can now become a valuation problem for two reasons:

1. To value one company, you have to value dozens:  Consider a firm with holdings in 25 other companies. Even if we could access information on these companies (because they are public), a thorough analysis of the firm would require a valuation of 26 companies. (Using the book value of these holdings, which are not marked to market, will yield skewed estimates. Using the market values of these holdings risks feeding any market mistakes into your valuation).
2. Some cross holdings cannot be valued: With many family group companies, some of the companies in the group are privately owned and never go public. As a consequence, there is little or no information that can be used to value companies. We have no choice but to use book value.

If you carry these concerns through to their logical conclusion, it is possible that investors either unconsciously (by using book value) or consciously (by discounting the market value of the cross holdings) will reduce the values of family group companies below what they would have been worth as independent companies. In effect, the sum of the parts will be greater than the whole. I have a paper on valuing cash and cross holdings that explores the technical details of this discount:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=841485

In my next post, I hope to examine the link between corporate governance and the phenomenon of cash and cross holdings.

Monday, May 17, 2010

"We are not in Kansas anymore"

It looks like that they have found the culprit for the 1000-point intraday swing on the Dow 30 on May 6. It turns out that rather than the hedge funds that were initially suspected, it was a Kansas-based money management firm, Waddell & Reed Financial, that traded 75000 e-mini S&P 500 contracts between 2.32 and 2.51 pm on May 6. That amounted to 9% of the trading volume on e-mini contracts during that period and all of the trading was executed by one trader at the firm. Incidentally, the CME report that uncovered this news also found that neither the trader nor the firm were acting imprudently or were at fault. This news item may still leave you a little bemused. How does one trader at a small money management firm cause a drop in market value of billions? And how can they not be at fault if they caused a market collapse? So, here is my attempt at providing an explanation.

What are e-mini contracts?
E-mini contracts are future contracts on the S&P 500. The "mini" in the name refers to the fact that each contract is $50 times the level of the index. (If the S&P 500 is at 1200, each contract is for a notional value of $60,000) E-mini contracts were introduced in the late 1990s by the Chicago Mercantile Exchange to provide a "smaller size" alternative to the long-standing S&P 500 futures contracts which were set to $500 times the level of the index; they have since been dropped to $250 times the level of the index.

How do investors use these contracts?
As with all futures contracts, there are speculators and hedgers in the e-mini market. The speculators buy or sell the e-mini to try to profit from overall market movements. Thus, a bullish (bearish) investor will buy (sell) e-mini contracts and make money if they are right on market direction. (If you buy 100 contracts and the S&P 500 moves up 80 points, you will make 100* 80 * 50 = $400,000; the 50 refers the fact that the futures contract is $50 times the index) The hedgers use the index to protect existing portfolio positions. Thus, a portfolio manager who wants to either protect profits already made or one who desires a floor on his or her losses will sell e-mini futures. (A portfolio manager who has $ 1 billion in equities can sell enough futures contracts to ensure that the value of the position does not drop below $ 900 million. Generally speaking, the more insurance you want, the more futures contracts you will have to sell. In more technical terms, you are creating a synthetic put on your portfolio, using options, and the number of futures contracts you will need to sell can be extracted using an option pricing model)

In many ways, the hedging position with futures is a lot more volatile than the speculative position, simply because the degree of selling is tailored to what the index does. As the index falls, the selling will often accelerate. partly because the point at which different portfolio managers hedge can vary. Thus, some portfolio managers may begin their hedging when the market drops 3%, others at 5% and still others at 10%.


How can futures affect the level of the index?
With financial futures, there is a third player that we have not mentioned in the section above, the arbitrageurs. Arbitrageurs have neither a market view nor do they have a portfolio to hedge. Instead, they are looking to make riskless profits, To prevent these profits, the futures price and the spot price are linked together in a rigid relationship:
Futures Price = Spot price (1 + riskfree rate - dividend yield)
To see why, assume that the S& P 500 is at 1000 right now, that the riskfree rate is 5% and that the dividend yield is 2%. Assume also that the one-year futures price on the index is 1045. Here is the arbitrage:
1. Borrow 1000 at the riskless rate and buy the index today at its spot price of 1000.
2. Sell the one-year futures contract at 1045.
3. During the next year collect dividends on the stocks in the index (2% of 1000 = 20). At the end of the year, deliver the stocks in the index in fulfillment of the futures contract and collect 1045. Pay the interest at the riskfree rate on the initial borrowing of 1000 (from step 1) and pocket the difference:
Profit = 1045 - 1000*.05 +20 = 15
To prevent this profit from occurring, the futures price has to be 1030. There are points at which you can quibble - being able to borrow at the riskfree rate and knowing the dividends for the next year - but they are minor ones, especially for the larger institutional players. The overall dividend yield on the S&P 500 index is very predictable and you can borrow at close to the riskfree rate, especially if you can back the borrowing up with marketable securities (as is the case here).

This futures-spot relationship creates the link. If one (spot or futures price) moves, the other has to follow. Thus, if there is an imbalance in the futures market, the futures price will change and the spot will follow. On May 6, here is how the script unfolded. The sell order placed by the trader at Waddell and Read was large enough to cause the e-mini futures price to drop significantly and the spot market had to follow. The fact that the trade was entirely driven by liquidity or hedging concerns (and not by information) resulted in a swift correction of both the spot and futures markets, with both reversing the losses by 3.30 pm.

What should be done about this?
I think that the May 6 collapse was an aberration. In what sense? The trade by the W&R trader occurred at a point in the day when the market was already skittish - it was down 250 points as worries about Greek default were rampant. When traders get antsy, they look for clues in trading by others. In other words, they assume that large trades, especially anonymous ones, must be coming from more informed traders (such as the Greek central banker) and they follow the trade.

While there is always the potential for this type of panic with or without futures markets, the existence of futures contracts has made it easier to create this type of panic. To the regulatory-minded, the solution seems simple. Ban futures trading or add more restrictions to the trading.  I disagree with the sentiment and think more harm than good will come out of it. As an investor who uses futures contracts very rarely and only to hedge, I still benefit from the liquidity created by these markets and bear little or no cost, simply because I choose not to trade frequently. In fact, as an intrinsic-value driven, long term investor, selling panics such as these can actually be opportunities to take positions in companies that I have always wanted to buy. For short term traders, though, futures markets may increase intraday volatility and thus their perception of risk in equities. I cannot speak for them but they are short term traders by choice!!