A reader (in fact, one that provided invaluable input into “Grow the Pie”) alerted me to the article “CEOs Are The Problem” by renowned MIT economist Daron Acemoglu. The article makes the following arguments:
- The current model of capitalism (shareholder primacy) is excessively influenced by Milton Friedman’s 1970 article that “The Social Responsibility of Business is to Increase Its Profits”.
- Capitalism has erroneously followed the recommendations of Michael Jensen – that CEO wealth should be more closely tied to shareholder value, to prevent “agency costs” that arise when CEOs pursue their own objectives rather than shareholders’.
- “The problem with shareholder primacy … was that it handed a massive amount of power to top managers. Many CEOs now run their companies according to their own personal vision. There is very little social oversight, and executive compensation has soared”.
- Under shareholder primacy, “there is nothing that prevents [CEOs] from pursuing excesisve automation to reduce labor costs, destroying jobs just to eke out a few more bucks for shareholders”.
- Shareholder primary allows excessive freedom to CEOs – manifesting in problems such as (3) and (4) above. The article’s final sentence is that “Corporate responsibility is too important to be left to corporate leaders”.
There is no doubt that Acemoglu is one of the most brilliant economists of his generation. During my time as an MIT PhD student, he was awarded the John Bates Clark medal for outstanding contributions to economics by a person under 40 (similar to a Fields Medal in mathematics) – a strong predictor of the Nobel Prize in Economics. His work has been cited 177,000 times, with a single paper (“The colonial origins of comparative development”) having 15,000 citations – nearly 4,000 more than all of life’s work put together. His influence extends far beyond academia, with books such as “Why Nations Fail” and “The Narrow Corridor”.
However, Acemoglu’s expertise lies in issues such as political economy and macroeconomics, which are a far cry from responsible business. As a result, his article contains basic misunderstandings and contradictions, learn more at this site.
For example, points (2) and (3) above are inconsistent. Jensen’s work highlights the problems that arise when managers have unlimited discretion, and stressing the importance of accountability. His 1990 paper with Kevin Murphy, which Acemoglu criticises, aims to prevent cases where “executive compensation has soared” by ensuring that CEOs are only well-paid if they create long-term shareholder value. Jensen’s 1976 paper with William Meckling – the most cited academic business article of all time – introduced the idea of agency costs. It’s often misinterpreted as defining agency costs as arising whenever managers don’t maximise short-term profit, and arguing that institutional forces should ensure such maximisation. But, as I explain in “What stakeholder capitalism can learn from Jensen and Meckling“, this article is widely misunderstood. Instead, it acknowledged that shareholders may have non-financial as well as financial objectives. Agency costs arise when a CEO fails to pursue shareholders’ financial objectives (e.g. due to coasting or failing to innovate) or non-financial objectives (e.g. exploiting workers, or prioritising her own non-financial goals such as donating to an opera house when shareholders are more concerned with labor rights).
Similarly, points (1) and (5) contain a major contradiction. Friedman’s article is often interpreted as claiming that the goal of society is to maximise short-term profit – that’s why business’s social responsibility is to maximise it. This article is also widely misinterpreted and I clarify it in “What stakeholder capitalism can learn from Milton Friedman.” Friedman agrees that “corporate responsibility is too important to be left to corporate leaders” – he stresses that much of the responsibility lies with governments. The context in which Friedman was writing (often ignored in criticisms of his article) was one in which governments were pressuring companies to take socially responsible actions, to make up for their own failures. But, the government has far more power to achieve social objectives than companies – using income taxes to address inequality, carbon taxes to tackle externalities, and spending to alleviate regional issues (e.g. unemployment in the North of England). Expecting companies to solve all of the world’s problems is placing unrealistic burdens on them. If a CEO has created long-term, sustainable value, it’s fair for the company to pay her handsomely – it shouldn’t be criticised for contributing to inequality.
Moreover, the whole article misunderstands the concepts of shareholder primacy and shareholder value, which I clarify in “Why shareholder capitalism benefits wider society“. Acemoglu argues that shareholder capitalism pressures CEOs to exploit society in the pursuit of short-term profit. This makes no sense, because shareholder value is an inherently long-term concept – Finance 101 highlights that it’s the present value of all future cash flows. A CEO pursuing long-term shareholder value will invest in her workers, not exploit them. It’s true that some CEOs who do the latter, but they are not pursuing shareholder value but instead short-term profit (potentially due to a poorly-designed incentive scheme). The solution is to tackle the specific problem (e.g. incentive redesign) rather than throw the baby out with the bathwater and reduce shareholder accountability.
The above contradictions are not just academic arguments – they matter for the real-world solutions. Despite my concerns with Acemoglu’s article, I agree with a less extreme version of the title – some CEOs are the problem. What this means is that Jensen’s concept of shareholder accountability is the solution, not the problem, and should be encouraged. Similarly, rather than being ridiculed, supporters of stakeholder capitalism can use Friedman’s article to guide us – by clarifying the different roles that governments and companies should play and not pushing the entire responsibility to repurpose capitalism on CEOs.
But isn’t there the concern that shareholder oversight – even if it focuses on long-term shareholder value – will nevertheless ignore stakeholders? Won’t it lead to the destruction of jobs, as per point (4) above? It may do – but the destruction of jobs may actually be good for society. Indeed, many technological advances that destroy jobs in the short-term have had hugely positive social impacts – such as the tractor and the ATM. Not only did this significantly improve productivity, but it also removed the need for humans to work on menial tasks that offer little skills development.
Moreover, many of the technological advances that we need to satisfy the world’s most pressing problems will erode jobs – moving to renewable energy will endanger the employment of 5 million employees in the Chinese coal sector. But, the goal of society is not actually to provide jobs. It’s to assign citizens to careers where they can use their talents to serve society and flourish as humans. Employing someone in a role that harms society (e.g. burning coal) or can be easily done by a machine (e.g. cheque deposits) is a massive waste of their time and talents. That doesn’t mean that companies can introduce technology with impunity; instead, they should reallocate employees to more value-adding tasks that cannot be replaced by machines. In his excellent TED talk, Acemoglu’s MIT colleague David Autor explains how the ATM allowed bank tellers’ jobs to move away from simply cashing in cheques to providing customer advice.
Indeed, contrary to common concerns that shareholder activism forces companies to focus on short-term profit, rigorous evidence shows that it creates long-term value for both shareholders and stakeholders. In contrast, when managers are insulated from shareholder pressure, managers pursue the quiet life and coast – failing to both close down old, unproductive plants and create new ones.
The Bigger Picture
Over and above the specific points in Acemoglu’s article, it’s an example of a more general concern. Responsible business is now a hot topic, which encourages everyone to jump on the bandwagon and write about it, regardless of expertise. How does a discerning reader know what to trust? My TED talk, “What to Trust in a Post-Truth World“, and cut-out-and-keep guide to Evaluating Research, provide some potential tips. Here are two:
- Critically analyse the author’s credentials (e.g. publications in top-tier academic journals) in the relevant field. This avoids the issue of “halo effects”, where someone’s expertise in one field leads to them being seen as an expert in many fields. Indeed, this is why the reader alerted me to this article – concerned that Acemoglu’s influence would mean that many readers would take his article at face value. Beyond this article, a visionary tech entrepreneur is currently see as the leading authority on climate change, even though climate science is a quite different topic.
- Ask whether the article is balanced. There are almost two sides to every issue in social science. Thus, while a one-sided title may grab headlines, it suggests that the article may not have seriously considered the evidence on both sides. “CEOs are the problem” is a sweeping over-generalisation. Some CEOs are the problem; others are not. Thus, moderate solutions may be warranted, even though extreme remedies might attract more attention.