Do Workers on Boards Protect Jobs in a Downturn?

A hot topic is whether companies should have workers on their boards, to ensure they act in the interest of stakeholders, not just shareholders. This debate was ongoing even before COVID-19 – for example, Theresa May initially suggested that she’d mandate worker representation when she became UK Prime Minister, but then decided against it. It’s of even greater relevance now. Mass layoffs will definitely occur due to COVID-19 and have already started happening. Worker representation might help to preserve jobs.

The suggestion that worker directors may improve corporate governance should be taken very seriously. However, since particular parties might have a vested interest in this issue (e.g. trade unions might be naturally predisposed to support it, and shareholder groups to oppose it), there’s a big danger that people quote “evidence” just because it happens to support their viewpoint, regardless of its rigor. This is a particular risk because it’s easy to attribute causation when the data only shows correlation. For example, I’ve seen claims that Germany’s superior economic performance is due to having workers on the board. However, Germany differs from the UK and US in many other ways. It would be equally premature to conclude that Germany’s superior economic performance is because its citizens speak German, and the UK and US should speak German to improve their performance.

A paper by E. Han Kim, Ernst Maug and Christoph Schneider indeed provides causal evidence of the effects of worker representation and suggests a potential benefit – it leads to employees being better protected against economic downturns (such as COVID-19, although of course their dataset predated COVID-19). Shareholders also benefit through workers being willing to accept lower wages in return for this insurance. Even though I’ve been skeptical on the benefits of worker representation, because the prior evidence was weak, I found this to be the most convincing paper in favor of worker representation. There are still twists though – it protects the jobs of white-collar and skilled blue-collar workers, but not unskilled blue-collar workers. This is because it’s more likely that representatives from the first two groups successfully get elected to the board.

Below follows a post on the Review of Finance Managing Editor’s blog by RF Advisory Editor Andrew Ellul, which describes the paper in a non-technical manner. (I adapted it slightly for a practitioner audience). The interested reader should also check out this article by LBS Executive Fellow Tom Gosling, where he summarises Ernst Maug’s presentation of his paper at the LBS Centre for Corporate Governance and then adds his own thoughts, setting the research in the wider context.

Governance to Enforce Implicit Contracts between Firms and Workers

There is a long-standing literature investigating the implicit contract between companies and workers, dating back at least to Knight (1921). The idea that is that companies are risk-neutral because their shareholders are diversified, but employees are risk-averse as their entire income comes from the company. Thus, companies should provide insurance to employees from downturns, by guaranteeing their jobs; in return, employees should be willing to accept a lower salary.

Examples of firms providing such employment insurance at times of economic stress have attracted attention from politicians and the popular press. In one significant part of his acceptance speech in 2012, President Barack Obama said:  “The family business in Warroad, Minnesota, that didn’t lay off a single one of their four thousand employees during this recession, even when their competitors shut down dozens of plants, even when it meant the owners gave up some perks and pay … understood their biggest asset was the community and the workers who helped build that business…

Two assumptions are needed for implicit contracts to work:

  1. The contract has to be self-enforcing – firms have to be able to commit to delivering the promise of stable employment, especially during bad shocks. As Han, Ernst and Christoph ask: “If workers accept lower wages today, how can shareholders commit not to lay them off in the future when it is in their best interest to do so?”
  2. Workers must value the employment and wage insurance provided by the firm.

This paper focuses on one self-enforcement mechanism: labor representation on the firms’ boards. The authors use plant-level data from Germany for one good reason: Starting in the 1970s, Germany was at the forefront of a movement granting more labor representation on boards. As the Figure below shows, most countries now grant their employees workers some form of representation, not always, though, at the board level; only the United States and Singapore are the two OECD countries that do not grant workers any form of representation. Thus, the benefits and costs of labor representation are important issues for a number of developed countries.

The authors exploit the parity-codetermination in the German system, whereby the law requires 50% employee representation on supervisory boards when firms have more than 2,000 employees working in Germany. This employment cut-off provides the laboratory used to investigate whether employment and wage insurance is provided and the consequent impact on firm performance.

Firms with Labor Representation Provide Employment Insurance…

The paper finds that two groups of workers, white-collar and skilled blue-collar workers, in parity-codetermined firms receive employment insurance: these workers are protected against layoffs during negative (industry-level) shocks. Such employment insurance is not received by workers in non-parity firms. Another group of workers, unskilled blue-collar workers of the same parity firms, receive no such employment insurance during such negative shocks. This is surprising since implicit contract theory suggests that all workers should be protected. The authors argue that this is because unskilled blue-collar workers are not represented on supervisory boards, reinforcing the conjecture that worker participation in governance is necessary to enforce implicit contracts.

The authors are aware that parity-codetermination may not only be an enforcement mechanism but it may also serve to entrench workers, by presenting efficient layoffs during bad economic conditions. Distinguishing between these two competing hypotheses is crucial for the correct interpretation of the evidence. The authors use wage differentials to achieve this objective: while the implicit contract theory implies lower wages as compensation for insurance, no such wage concessions are contemplated by the worker entrenchment hypothesis.

…but Pay Lower Wages and Realize No Shareholders’ Gains

The paper finds that workers with vocational and higher educational qualifications, the two categories from which most skilled blue-collar and white-collar workers are drawn, receive 3.3% lower wages at parity-codetermined firms. The authors interpret this wage concession as a quid pro quo in exchange for employment insurance.

These two results – the cost to shareholders arising from employment insurance and the benefit of lower wages – give rise to a central question: do shareholders benefit from providing insurance? The paper finds that parity firms’ operating performance and valuation suffer more during shock periods compared to non-parity firms. So, it’s important to ask: is the wage concession enough to overcome the negative valuation impact in downturns?

To answer this question, the authors analyze parity firms’ performance and shareholder value through the entire business cycle – over the non-shock and shock periods – and find no impact arising from parity codetermination. On average, parity firms perform no worse or better than nonparity firms. This result implies that any efficiency gains from risk-sharing are mostly captured by workers, and generate no gains to shareholders. This is why firms may not voluntarily adopt labor representation in firm’s governance mechanisms, and so a regulatory intervention is needed to introduce such representation. But overall, it seems that the pie grows – workers’ slice rises and shareholders’ slice doesn’t fall.