Are conglomerates easy or difficult to analyze? On the one hand, their diversified business lines should balance each other, making the group less cyclical and smoothing its overall performance. On the other, they will test an investor’s ability to understand the company and value it accurately.
Consider two giants that reported results earlier this month. Warren Buffett’s Berkshire Hathaway revealed that its railroad unit had increased its profits by 24%, while its insurance arm slumped to an underwriting loss of $22 million, compared with a gain of $337 million a year earlier.
Disney, meanwhile, enjoyed higher than expected income from its theme parks and resorts around the world, but was penalized by the market for troubles at ESPN.
However, the intricacies of mastering both freight dynamics and insurance modeling or international tourism and domestic media consumption do not seem to deter portfolio managers from owning these stocks. Disney is owned by more than 3,000 funds and Berkshire B stock is held by more than 2,500.
In July, Dallin Alldredge of Washington State University published a study investigating whether investors were wise to buy conglomerates or whether they would be better off sticking to simpler businesses.
Alldredge defined conglomerates as firms with business operations in multiple industries, as indicated by the Standard Industrial Classification codes of their revenue streams. That was in contrast to ‘standalone’ firms with operations in just one industry.
As you would expect, the average conglomerate was 87% larger than the standalones by market capitalization, and was also typically older and less volatile. Conglomerates also offered a higher dividend yield of 3.96%, compared with 2.45% from simpler corporations.
This perhaps explains why conglomerates are more popular with professional investors: although the average institutional ownership of both conglomerates and standalones was very similar – at around 50% each – Alldredge observed that ‘given that there are more than twice as many standalone firms than conglomerate firms in the universe of stocks available to financial institutions, the near parity between the percentage of conglomerate-firm stock and standalone-firm stock in the portfolios of financial institutions suggests a preference for conglomerate-firm stocks’.
Using 13F disclosures and excluding passive investors, Alldredge wondered next whether the fund managers were able to trade these conglomerates more profitably than standalone companies. He discovered they were not.
Alldredge constructed long/short portfolios, buying either the conglomerates or the standalones that were the most heavily purchased by financial institutions over the previous quarter, and shorting those stocks in each category that had been most the heavily sold.
The conglomerate portfolio generated profits that were insignificantly different from zero, but the standalone portfolio produced abnormal returns of 65 basis points per month. In other words, the institutional investors added value with their activity in simple stocks but not in complex ones.
Moreover, the success of the institutions’ trading gradually declined as they dealt in more sprawling entities. When the conglomerate s operated across just two industries, for instance , the portfolios yielded excess returns of 26 basis points per month. This dropped into negative alpha territory for conglomerates with operations in t hree industries, and excess losses widened to 18 basis points a month for conglomerates in more than three sectors.
For Alldredge, the explanation was that asset managers could not adequately analyze multi-faceted conglomerates, even though they did seem able to do so for more streamlined stocks.
‘These findings provide unique and direct evidence suggesting that attention scarcity limits investors’ ability to process information,’ Alldredge concluded.
‘Institutions with high conglomerate concentration portfolios are attention-constrained and unable to allocate the necessary attention to the conglomerate firms in their portfolios to effectively utilize their information-processing skill, which hinders investment performance.’
So when faced with a portfolio manager who owns the likes of Berkshire or Disney, it may be worth enquiring about their thoughts on the former’s control of Duracell or the latter’s pipeline of Star Wars products. If they don’t have an opinion, a manager who focuses on more straightforward stocks may be a better option.