The investing world would be a simpler one (albeit with less opportunities), if all listed companies just owned a single operating business without any other significant financial interests or non-operating assets. In reality, a large number of listed companies either operate multiple businesses or have non-operating assets and financial interests accounting for a sizeable portion of their value.
In my article "How Benjamin Graham Will Possibly Invest In A World Without Net-Nets," I highlighted that the number of U.S.-listed net-nets has declined significantly, compared to the time when Benjamin Graham first introduced the concept to the investment community. One of the contributing factors is that sophisticated screening tools have allowed investors to arbitrage the net-net investment opportunities.
In comparison, stocks trading at discounts to their sum-of-the-parts valuations are still in abundance today, because such opportunities mostly have to be discovered by hand, assessing the value of disparate assets and businesses one by one. In addition, sum-of-the-parts discounts tend to exist for stocks across the market capitalization spectrum, making this value investing strategy much less liquidity-constrained than that of net-nets. Readers with a larger amount of capital to deploy can also refer to my other article "A Case Study On Large-Cap Value Investing."
I view net cash stocks, companies with net cash (cash and short-term investments net of all interest-bearing liabilities) accounting for a significant percentage of their market capitalization, as sum-of-the-parts discounts as well. But I have already covered this area of investment opportunities in the article mentioned above.
Before even attempting to put a price tag on the individual parts of a holding company, one needs to first ensure that the parts are indeed discrete and can be independently valued without overlapping with the other parts of the company. For example, a company vertically integrated across the value chain might see its downstream business become less valuable if it sells its upstream business as it loses access to low-cost internally sourced raw materials. Another example will be a company operating on a razor-and-blade model, where its maintenance & repair business will suffer from a stagnant or declining installed base from its "core razor business."
After establishing the separate parts of a listed holding company that can be valued separately, things don't necessarily get easier. For holding companies owning multiple businesses, the typical approach is to value the respective listed subsidiaries & associates at their current market capitalizations and apply appropriate multiples to the earnings or net asset values of the privately-held entities.
It is possible to get the value of any single business wrong and that error get multiplied many times in a sum-of-the-parts valuation. Similarly, appraising the value of hidden real estate (probably carried at historical cost on the books) is an art rather than science, with the estimated value of properties varying widely depending on the choice of cap rates and the health of the individual real estate segments.
Another form of hidden assets is Net Operating Losses, where the timeline and quantum of the actual tax benefits realized are largely a judgment call. Therefore, the wider the valuation gap between the parts and the sum (listed holding company's market price), the greater the margin of safety an investor enjoys for such sum-of-the-parts discounts situations.
I am likely quoting Aristotle out of context here, but when he said that "The whole is greater than the sum of its parts," this does not necessarily hold true in the investment world. When individual businesses or assets are connected together as one single listed entity, they are usually not worth more and the holding company is accorded a "conglomerate discount" by the market.
Professor Aswath Damodaran takes the view that multi-business conglomerates are deserving of a "complexity discount" and he came up with a complexity scoring system (table below) to relate a company's complexity to the size of the discount that it deserves. Damodaran used a regression to come up with an equation to relate a company's P/B ratio to its expected growth rates, betas, ROEs and number of pages in its 10-K.
The regression equation he derived for his January 2006 research paper was P/B = 0.65 + 15.31 ROE - 0.55 Beta + 3.04 Expected growth rate - 0.003 multiplied by the number of pages in 10-K, implying that a firm with a 15% return on equity, a beta of 1.15, and expected growth rate of 10% and 350 pages in the 10-K would have a price to book ratio of 1.54.
Most investors, including myself, probably feel that the derivation of the complexity discount is a tad too complex. Unfortunately, the common alternative of simply applying an arbitrary discount isn't the most ideal solution as well.
Coming back to Aristotle's quote, the whole can be greater than the sum of the parts, if the holding company has a superb capital allocator at the helm with the ability to drive value creation. One simple illustration of this is the classic growth-share matrix pioneered by the Boston Consulting Group. A good capital allocator can direct excess capital generated by cash cows (high market share, low growth potential) to stars (high market share, high growth potential). In theory, a well-run holding company can enjoy the benefits of size and scale by virtue of better (and cheaper) access to financing from banks and superior valuations due to a larger market capitalization putting them on the radar of more institutional investors.
The Narrowing Of The Gap
While many value investors, including myself, are firm believers that undervaluation itself is the best catalyst, sum-of-the-parts discounts tend to persist for a longer period of time in comparison with other deep value situations like net-nets where cheapness is rather obvious. Events like recapitalizations, share buybacks, special dividends, disposal of non-core assets and planned spin-offs are likely to precede any actual narrowing of the gap between the sum-of-the-parts valuation and the market price.
Sum-of-the-Parts Discounts Case Studies
Tiffany & Co. (NYSE:TIF)
Apart from net-nets and the Magic Sixes (companies trading at less than 0.6 times P/B, 6 times P/E and with dividend yields of 6% or more), Peter Cundill, a Canadian investing legend and one of the great deep value investors of all time, invested in sum-of-the-parts discounts with hidden assets on and off the balance sheet. In the Peter Cundill biography authored by Mr. Christopher Risso-Gill, "There's Always Something to Do: The Peter Cundill Investment Approach," a couple of Cundill's past investments were shared, including Tiffany.
In the 1970s, Cundill bought Tiffany shares at an average price of $8 per share (below book value) and exited his position completely at $19 within a year. Tiffany's hidden assets on and off the balance sheet included: (1) a 128.5 carat canary-coloured Tiffany Diamond carried on the company's books at $1.00 but was worth more than $2 million, (2) its Fifth Avenue store (freehold Prime Manhattan real estate) carried on the company's books at a 1940 historical cost of $1 million, and (3) the brand value of Tiffany (which was totally off-balance sheet since the company had zero goodwill). Six months after Cundill's sale of Tiffany shares or more specifically in November 1978, Tiffany was sold to Avon Products (NYSE:AVP) at $50 per share. This is a further validation of the sum-of-the-parts discount at Tiffany, which Cundill discovered and took full advantage of.
I also wrote about Tim McElvaine's past successful sum-of-the-parts discount investment in Howard Hughes (NYSE:HHC) in an article titled "Peter Cundill's Shopping List For U.S. And Asian Magic Sixes." McElvaine worked at Peter Cundill & Associates Ltd. for 12 years together with Peter Cundill, before starting his own firm, McElvaine Investment Management.
Poongsan Holdings (005810 KR)
I first learnt about Korea-listed holding company Poongsan Holdings from a July 2013 Value Investor Insight interview with Korean asset management firm Petra Capital Management. Poongsan Holdings is a holding company with multiple operating subsidiaries whose most significant holding is its 33% interest in Poongsan Corporation (103140 KR), a manufacturer of fabricated non-ferrous metal products which Poongsan Holdings spun off in 2008. As per the chart below (the Y-axis is the market capitalization or value in millions of South Korean Won), Poongsan Holdings (blue line) used to trade significantly below the market value of its 33% interest in Poongsan Corporation (red line), but the sum-of-the-parts discount was gradually narrowed over time and eventually closed in late 2014.
Asia/U.S. Deep-Value Wide-Moat Stocks Premium Research
Subscribers to my Asia/U.S. Deep-Value Wide-Moat Stocks exclusive research service get full access to the list of deep-value & wide moat investment candidates and value traps, including "Magic Formula" stocks, wide moat compounders, hidden champions, high quality businesses, net-nets, net cash stocks, low P/B stocks and sum-of-the-parts discounts.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.