My first exposure to "conglomerates" was not calculated to attract my sustained interest as an investor. As a young lawyer, I worked on the National Student Marketing litigation (Andrew Tobias has a great book on this - "The Funny Money Game"). In the early 1970's, a number of growth conglomerates appeared. These companies would combine a variety of unrelated businesses (e.g. a movie studio, a defense contractor, a fine arts auction house, a fertilizer producer, and a maker of toilet seats) and the investment community would fall in love with the idea. It was never exactly clear what the attraction was - in some cases, the notion was that the CEO was a "genius" who could turn each of these businesses into gold. In fact, what was often going on was the perpetuation of "earnings growth" through the acquisition in stock-for-stock deals of companies whose PE multiples were lower than the conglomerate's. If company A trades for 30 times earnings and it acquires company B in a stock for stock deal at a PE of 15 , then company A's per share earnings automatically increase. A kind of "momentum" was developed and more and more inventive arguments were made to justify less and less strategic acquisitions. Needless to say, it did not come to a good end.
I think that, for a while, this may have poisoned the water and there were frequently instances of conglomerates trading at bargain levels. Companies which operate a variety of businesses are hard to analyze. Metrics like margin, free cash flow, and return on investment can be meaningless when applied to a financial statement which consolidates the results for a company which includes restaurants, oil drillers, insurance subsidiaries and newspapers. It is also virtually impossible to value these companies by comparing them with "comparables" because there really are no comparable combinations of disparate businesses.
This is where "sum of the parts" analysis comes in. In certain instances, an investor can attempt to calculate the value of the sum of the parts of one of these companies and then compare it with the market cap to determine whether there is a bargain to be had. This is much easier when the "parts" are shares in publicly traded companies. Theoretically, all you have to do is calculate the number of shares held by the holding company and multiply it by the latest market price. In certain cases, an investor can determine that the sum of the publicly traded parts is large enough that a buyer of the stock is getting the rest of the company for free or at a huge discount.
One such company on which I am bullish is Loews (L). It owns a large controlling interest in Diamond Offshore (DO), Boardwalk Pipeline (BWP), and CNA Insurance (CNA), each of which is publicly traded, allowing the value of L's position to be evaluated. On top of this, it has a nice wad of net cash, owns Loews hotels and also owns High Mount Exploration (an oil and gas exploration and production company). This story has been written up frequently and a recent sum of the parts analysis is available on the L website. When you add up the value of the DO, BWP and CNA positions and throw in the cash, you get to about $40 a share (L closed today at $42.02) and it turns out that you are just about getting the hotels and High Mount for free.
I recently did a story about a Business Development Company - Capital Southwest (CSWC). It owns big positions in some publicly traded companies and also has positions in privately held companies. When you calculate the value of the stock in the public companies and add in CSWC's net cash, you are getting the interests in the non-public companies for 15-16% of book value(which is supposed to reflect fair market value).
Washington Post (WPO) is another interesting situation. It owns Kaplan Education, the newspaper, Newsweek, some cable operations and some radio and TV stations. Valuation is less precise because an investor does not have a public market price to use to gauge the value of the "parts," but some fairly conservative assumptions can be made and this stock still looks cheap. Of course, a great deal depends upon how Kaplan and the for-profit education sector do going forward.
I am long all three of these but certain caveats should be taken into account.
- The Quality and Integrity of Management is Vitally Important. Sum of the parts situations will not create true value for shareholders unless management is committed to shareholder friendly policies and a philosophy of management which reflects the manager's fiduciary duty to the shareholders. It is very easy for management to abuse the situation through excess compensation or self-dealing(including the sale of one of the parts at less than fair value to "friendly" investors). In the case of L, the Tisch family has an excellent and well-deserved reputation for looking out for shareholder interest. CSWC's management appears to have a similar orientation and track record. WPO's management is struggling with a difficult transition in the print media area but it also appears to have its eye on the ball with respect to education and other businesses. On the other hand, the Tyco saga provides a cautionary lesson in how a stumbling management can undermine shareholder value.
- Analyze the Parts - When you invest in one of these, you are taking positions in each of the parts and you should develop a sense of whether these positions make sense for your portfolio. I lost money in my position in WPO because, even though I was getting the "parts" at a discount, the largest "part" (Kaplan) got hammered so badly that it tanked WPO in the market.
- Be Patient - These stocks are undervalued on a sum of the parts basis today and they are likely to be undervalued on a sum of the parts basis tomorrow. It may take years for a value unlocking "catalyst" to emerge.
One of the most difficult questions in this area is exactly how widely to apply sum of the parts analysis. A case like L is easy because the positions are in publicly traded companies that function with a good deal of autonomy. But many large companies are made up of diverse parts whether they are organized into subsidiaries or divisions and the valuation of these parts may suggest the conclusion that stock is undervalued. However, the "catalyst" may be very remote indeed and, for a variety of reasons, the parts may become worth less when separated from the whole. Sum of the parts analysis is not a "silver bullet" for finding value stocks and has limited usefulness when applied to large integrated operations which can not be readily broken up. While not talismanic, it is certainly one of the important tools an investor should use in searching for value