In November, the US industrial powerhouse General Electric (GE) announced that it would split into three companies. The Japanese conglomerate Toshiba and healthcare giant Johnson & Johnson have since announced similar plans. And these are just the latest in a string of such breakups, which include the likes of DowDuPont and Siemens. Is the age of the conglomerate coming to an end?
While tech companies such as Alphabet (Google’s parent company), Amazon, and Meta (formerly Facebook) focus on acquisitions, conglomerates increasingly view breaking up and streamlining their businesses as a way to improve performance. GE stock, for example, has been underperforming for years. Shareholders are betting that its healthcare, aviation, and energy divisions will be able to reap higher profits and compete more effectively over the next century if they can set their own paths, with more focused business models, tailored capital allocation, and strategic flexibility.
This approach seems to have worked for Siemens, which credits deconglomeration with helping it to overtake GE, its historical rival, and post healthy profits for 2021. But the question remains: Why are conglomerates underperforming? The answer comes down to the so-called conglomerate discount. Markets tend to undervalue the stock of companies with a diversified set of businesses. If each of their divisions was valued at the level of a single-focus firm in their sector, they would be worth more overall.
The conglomerate discount reflects the fact that conglomerates have lower Tobin’s Q – the ratio between a physical asset’s market value and its replacement value – than single-product firms. This is partly because capital is often misallocated across divisions. When headquarters over-allocate to some divisions, the firm’s return on assets declines.
The conventional wisdom is that this misallocation of capital is a result of “corporate socialism” – when head offices allocate capital from more profitable to less profitable divisions, in order to give the underperforming division the resources and incentives to improve. But, as my co-authors and I show in recent research, that is not the whole story.
In fact, head offices are more likely to over-allocate capital to a conglomerate’s best-performing divisions. That is because, in deciding where to send more capital, they review proposals from division managers. Those divisions that promise to bring in the most profits get funded, with the specific amount determined by the estimated costs put forward in the proposal.
But managers are empire-builders, with a strong appetite for running larger divisions, so they often inflate the estimated costs of their projects. If they are already running the conglomerate’s more profitable divisions, the excess costs will not be enough to prevent headquarters from handing over the funds. Nonetheless, this arrangement obviously does not imply the most efficient use of capital – hence the conglomerate discount.
Yet abandoning the conglomerate model is hardly the only way to escape this trap. More market competition can go a long way toward curbing division managers’ empire-building behaviour, because it intensifies pressure on headquarters to allocate funds toward the most efficient projects.
This is not mere speculation. In our research, my co-authors and I find that since China joined the World Trade Organization 20 years ago, US conglomerates facing competition from Chinese firms have benefited from significant reductions in over-reporting of costs by managers. Over-reporting dropped by 15% per standard deviation increase in Chinese imports, based on 7% average annual growth in imports from China between 1999 and 2007.
Costs fell the most in the companies’ most productive business segments, which had previously been over-reporting by the largest margin. These segments were then able to increase their output, while less productive business segments continued to underperform, or even contracted. This meant that the most efficient divisions received the most capital.
These changes led to a sharp decline in firms’ conglomerate discount – a 32% reduction, per standard deviation increase in Chinese imports. After the 2008 global financial crisis, exporting conglomerates – for which competition from China was strongest – even ended up securing a “conglomerate premium”. Meanwhile, conglomerates that did not face competition from China saw their conglomerate discount grow.
Superstar firms often have so much market power that they can raise prices without losing many customers, meaning that they often end up delivering too little output at too high a price. By challenging their monopoly power, China countered these market distortions, compelling conglomerates to increase output and reduce prices.
Firms like GE and Siemens may well find that their decision to split their businesses into multiple companies leads to increased profits and higher stock prices. But other ailing conglomerates might find that all they really need is a healthy dose of competition.
Dalia Marin is a professor of international economics at the School of Management of the Technical University of Munich, a research fellow at the Centre for Economic Policy Research and a non-resident fellow at Bruegel.
Copyright: Project Syndicate